H44-A2 Vol. 24 No. 8 Replaces H44-A Vol. 17 No.

15

Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition

This document provides guidance for the performance of reticulocyte counting by flow cytometry. It includes methods for determining the trueness and precision of the reticulocyte flow cytometry instrument and a recommended reference procedure. A guideline for global application developed through the NCCLS consensus process.

NCCLS...
Serving the World’s Medical Science Community Through Voluntary Consensus
NCCLS is an international, interdisciplinary, nonprofit, standards-developing, and educational organization that promotes the development and use of voluntary consensus standards and guidelines within the healthcare community. It is recognized worldwide for the application of its unique consensus process in the development of standards and guidelines for patient testing and related healthcare issues. NCCLS is based on the principle that consensus is an effective and costeffective way to improve patient testing and healthcare services. In addition to developing and promoting the use of voluntary consensus standards and guidelines, NCCLS provides an open and unbiased forum to address critical issues affecting the quality of patient testing and health care. PUBLICATIONS An NCCLS document is published as a standard, guideline, or committee report. Standard A document developed through the consensus process that clearly identifies specific, essential requirements for materials, methods, or practices for use in an unmodified form. A standard may, in addition, contain discretionary elements, which are clearly identified. Guideline A document developed through the consensus process describing criteria for a general operating practice, procedure, or material for voluntary use. A guideline may be used as written or modified by the user to fit specific needs. Report A document that has not been subjected to consensus review and is released by the Board of Directors. CONSENSUS PROCESS The NCCLS voluntary consensus process is a protocol establishing formal criteria for: • • • • the authorization of a project the development and open review of documents the revision of documents in response to comments by users the acceptance of a document as a consensus standard or guideline. the need for field evaluation or data collection, documents may also be made available for review at an intermediate (i.e., “tentative”) consensus level. Proposed An NCCLS consensus document undergoes the first stage of review by the healthcare community as a proposed standard or guideline. The document should receive a wide and thorough technical review, including an overall review of its scope, approach, and utility, and a lineby-line review of its technical and editorial content. Tentative A tentative standard or guideline is made available for review and comment only when a recommended method has a well-defined need for a field evaluation or when a recommended protocol requires that specific data be collected. It should be reviewed to ensure its utility. Approved An approved standard or guideline has achieved consensus within the healthcare community. It should be reviewed to assess the utility of the final document, to ensure attainment of consensus (i.e., that comments on earlier versions have been satisfactorily addressed), and to identify the need for additional consensus documents. NCCLS standards and guidelines represent a consensus opinion on good practices and reflect the substantial agreement by materially affected, competent, and interested parties obtained by following NCCLS’s established consensus procedures. Provisions in NCCLS standards and guidelines may be more or less stringent than applicable regulations. Consequently, conformance to this voluntary consensus document does not relieve the user of responsibility for compliance with applicable regulations. COMMENTS The comments of users are essential to the consensus process. Anyone may submit a comment, and all comments are addressed, according to the consensus process, by the NCCLS committee that wrote the document. All comments, including those that result in a change to the document when published at the next consensus level and those that do not result in a change, are responded to by the committee in an appendix to the document. Readers are strongly encouraged to comment in any form and at any time on any NCCLS document. Address comments to the NCCLS Executive Offices, 940 West Valley Road, Suite 1400, Wayne, PA 19087, USA. VOLUNTEER PARTICIPATION Healthcare professionals in all specialties are urged to volunteer for participation in NCCLS projects. Please contact the NCCLS Executive Offices for additional information on committee participation.

Most NCCLS documents are subject to two levels of consensus—“proposed” and “approved.” Depending on

Volume 24 Number 8

H44-A2 ISBN 1-56238-527-5 ISSN 0273-3099

Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition
Charles F. Arkin, M.D., Chairholder Bruce H. Davis, M.D., Vice-Chairholder J. David Bessman, M.D. Berend Houwen, M.D., Ph.D. Frank M. LaDuca, Ph.D. Ginette Y. Michaud, M.D. Onno W. van Assendelft, M.D., Ph.D.

International Council for Standardization in Haematology (ICSH) Expert Panel on Cytometry George G. Klee, M.D., Ph.D., Chairholder Guiseppe d’Onofrio, M.D., Secretary Brian S. Bull, M.D. Ahnond Bunyaratvej Mauro Buttarello, M.D. Bruce H. Davis, M.D. Keiji Fujimoto, Ph.D. Warren Groner John A. Koepke, M.D. Jolanta Kunicka, Ph.D. S. Mitchell Lewis, M.D. Samuel J. Machin, MB, ChB, FRCPath, FRCP John Maples, Ph.D. R. Martin Rowan, M.D. Noriuki Tatsumi, M.D. Onno W. van Assendelft, M.D., Ph.D. Luc van Hove, M.D., Ph.D. Robert Magari, Ph.D.

Abstract
NCCLS document H44-A2—Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition provides guidance for the performance of reticulocyte counting by flow cytometry and automated hematology instruments. This guideline addresses methods for determining the precision and trueness of the flow cytometer and blood cell counters based upon principles of focused flow dynamics, along with recommendations for calibration and quality control. A description of the new methylene blue (NMB) method, a method against which the test instrument can be compared, is also included. Additional topics discussed include reference intervals and use of related reticulocyte parameters, i.e., the immature reticulocyte fraction (formerly termed “reticulocyte maturation index”) and reticulocyte hemoglobin content (CHr). NCCLS. Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition. NCCLS document H44-A2 (ISBN 1-56238-527-5). NCCLS, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA, 2004.

THE NCCLS consensus process, which is the mechanism for moving a document through two or more levels of review by the healthcare community, is an ongoing process. Users should expect revised editions of any given document. Because rapid changes in technology may affect the procedures, methods, and protocols in a standard or guideline, users should replace outdated editions with the current editions of NCCLS documents. Current editions are listed in the NCCLS Catalog, which is distributed to member organizations, and to nonmembers on request. If your organization is not a member and would like to become one, and to request a copy of the NCCLS Catalog, contact the NCCLS Executive Offices. Telephone: 610.688.0100; Fax: 610.688.0700; E-Mail: exoffice@nccls.org; Website: www.nccls.org

Number 8

NCCLS

This publication is protected by copyright. No part of it may be reproduced, stored in a retrieval system, transmitted, or made available in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise) without prior written permission from NCCLS, except as stated below. NCCLS hereby grants permission to reproduce limited portions of this publication for use in laboratory procedure manuals at a single site, for interlibrary loan, or for use in educational programs provided that multiple copies of such reproduction shall include the following notice, be distributed without charge, and, in no event, contain more than 20% of the document’s text. Reproduced with permission, from NCCLS publication H44-A2—Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition (ISBN 1-56238-527-5). Copies of the current edition may be obtained from NCCLS, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898, USA. Permission to reproduce or otherwise use the text of this document to an extent that exceeds the exemptions granted here or under the Copyright Law must be obtained from NCCLS by written request. To request such permission, address inquiries to the Executive Director, NCCLS, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898, USA. Copyright ©2004. The National Committee for Clinical Laboratory Standards.

Suggested Citation
(NCCLS. Methods for Reticulocyte Counting (Automated Blood Cell Counters, Flow Cytometry, and Supravital Dyes); Approved Guideline—Second Edition. NCCLS document H44-A2 [ISBN 1-56238-5275]. NCCLS, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania 19087-1898 USA, 2004.)

Proposed Guideline
November 1993

Approved Guideline
October 1997

Approved Guideline—Second Edition
February 2004

ISBN 1-56238-527-5 ISSN 0273-3099

ii

Volume 24

H44-A2

Committee Membership
Area Committee on Hematology
Charles F. Arkin, M.D. Chairholder Boston University Medical Center Boston, Massachusetts Bruce H. Davis, M.D. Vice-Chairholder Maine Medical Center Research Institute Scarborough, Maine J. David Bessman, M.D. University of Texas Medical Branch Galveston, Texas Berend Houwen, M.D., Ph.D. Beckman Coulter, Inc. Brea, California Frank M. LaDuca, Ph.D. International Technidyne Corporation Edison, New Jersey Ginette Y. Michaud, M.D. Center for Devices and Radiologic Health Food and Drug Administration Rockville, Maryland Onno W. van Assendelft, M.D., Ph.D. Centers for Disease Control and Prevention Atlanta, Georgia Advisors Dorothy M. Adcock, M.D. Esoterix Coagulation Aurora, Colorado Eugene L. Gottfried, M.D. Orinda, California J. Heinrich Joist, M.D., Ph.D. St. Louis University Health Sciences Center St. Louis, Missouri John A. Koepke, M.D. Durham, North Carolina Francis Lacombe, M.D., Ph.D. Hôpital Haut-Lévêque Pessac, France Jack Levin, M.D. VA Medical Center San Francisco, California Samuel J. Machin, MB, Ch.B, FRCPath, FRCP The University College London Hospitals London, United Kingdom Richard A. Marlar, Ph.D. Denver VA Medical Center Denver, Colorado Diane I. Szamosi, M.A., M.T.(ASCP)SH Greiner Bio-One, VACUETTE North America, Inc. Monroe, North Carolina Luc Van Hove, M.D., Ph.D. Abbott Laboratories Abbott Park, Illinois Staff Tracy A. Dooley, M.L.T.(ASCP) Staff Liaison NCCLS Wayne, Pennsylvania Jennifer K. McGeary, M.T.(ASCP), M.S.H.A. Project Manager NCCLS Wayne, Pennsylvania Donna M. Wilhelm Editor NCCLS Wayne, Pennsylvania Melissa A. Lewis Assistant Editor NCCLS Wayne, Pennsylvania

iii

Number 8 NCCLS iv .

...........26 Reticulocyte Hemoglobin Content (CHr)........................................................................................................................................................................................................................................................................1 Principle ...................................................................................2 10................7 Specifications for Automated Blood Counters and Flow Cytometers to Be Used for Reticulocyte Counting .........................4 5........................25 10..........................................................................................................................................................................................................3 6........11 Sample Preparation for Flow Cytometry....................................................................................................5 Comparability (Correlation) Method..........2 Definitions ........................................................................................................................2 6..........................5 5..................20 Testing for Carry-Over ...............................................23 Long-Term Quality Control Procedures.....................2 Specimen Collection and Storage ................................................5 6 6.............................................................................................i Committee Membership...........7 Counting Procedure ..............................................................................................................3 7.................................................................................................................................5 Specimen Collection Techniques ..............................1 10.............13 8...................24 Reference Population.6 Sample Evaluation....................1 6..............................................................................................................................................................3 10.......................................................................................11 Instrument Settings and Calibration ...................................................................1 Introduction................................................................................................................................4 7 Patient Information .................................................................................................................10 Correlation Method—New Methylene Blue Visual Microscopy ................................................................................................................................. vii 1 2 3 4 5 Scope......................25 Reticulocyte Reference Intervals................................................................................................................................................1 9........................................................5 Sample Integrity ...............................................4 8........................................................................................................................................................................................................................23 9..................................7 Slide Preparation ........................................................................................................................................................................................2 5.....................................7 Stain.......................26 Immature Reticulocyte Fraction (Reticulocyte Maturation Index).....................3 5...22 Short-Term Quality Control Procedures......................................................18 Specificity of Testing.............1 8...........................................................................19 Sensitivity of Testing........................................................................................8 Calculations ..................................27 v 9 Quality Control and Quality Assurance ....................11 Method(s) of Operation ............6 Specimen Storage ..................................1 5.....4 ........13 Internal Consistency Testing ....................................................................2 10 Reticulocyte Reference Intervals and New Reticulocyte Parameters .....12 8 Performance Testing of Automated Blood Counters and Flow Cytometers to Be Used for Reticulocyte Counting .............................................3 8................................................................Volume 24 H44-A2 Contents Abstract .............1 7...4 Reportable Range (Measuring Range)............. iii Foreword .....................2 8.......11 7.............................................................................2 7.......

.......................................................................................................................................34 Related NCCLS Publications..............................................................................................................................30 The Quality System Approach.....................................................28 Summary of Delegate Comments and Area Committee Responses ............................................Number 8 NCCLS Contents (Continued) References............35 vi ...............................................

Volume 24 H44-A2 Foreword The reticulocyte count is one of the last common hematology measurements still being performed using manual/visual methods. In addition. and Reference measurement procedure is combined with Reference method when referring to a thoroughly investigated measurement procedure shown to have an uncertainty of measurement commensurate with the intended use. H44-A2 was developed through the cooperation of the NCCLS Area Committee on Hematology and the International Council for Standardization in Haematology (ICSH). A Note on Terminology NCCLS. Harmonization is a process of recognizing. The inter. flow cytometric methods for reticulocyte counting have been shown to be more precise than manual/visual methods. the subcommittee has chosen to include the following ISO terms parenthetically in the text with the U. This manual/visual method may also be used as the back-up method or as the primary reticulocyte counting method in laboratories that do not have the preferred flow cytometric method available. for the automated reticulocyte flow cytometric methods. and explaining differences while taking steps to achieve worldwide uniformity. terms: Measuring range is combined with Reportable range when referring to a set of values of measurands for which the error of a measuring instrument is intended to lie within specified limits.and intralaboratory variables of these flow cytometric methods are becoming better understood. In light of this. For the sake of introduction and to avoid confusion. quality control/assurance methods are given that help ensure that the procedure is being performed with precision and trueness. NCCLS recognizes that medical conventions in the global metrological community have evolved differently in the United States. refers to the closeness of agreement between the result of a (single) measurement and a true value of a measurand. is committed to achieving global harmonization wherever possible. and that legally required use of terms. especially in assessing the trueness of other measurement procedures for the same quantity and in characterizing reference materials. ISO. These automated methods also afford the ability to determine other reticulocyte-specific parameters. the area committee and the expert panel believe that this guideline avoids duplication and advances the international harmonization of this important hematology measurement. if required. such as the immature reticulocyte fraction (IRF) based upon RNA content of these cells which further improve the diagnostic evaluation of anemic patients. regional usage. and CEN documents. By the cooperative development of a single document. thus comprising both random and systematic effects. In addition. that these differences are reflected in NCCLS. Implementation of this policy is an evolutionary and educational process that begins with new projects and revisions of existing documents. vii . and elsewhere. understanding. Europe. In keeping with NCCLS’s commitment to align terminology with that of ISO. NCCLS recognizes that harmonization of terms facilitates the global application of standards and is an area of immediate attention. The new methylene blue method outlined in the guideline serves as the comparative procedure. the following terms are used in H44: Accuracy in its metrological sense. Flow cytometric methods have been developed that significantly reduce the time it takes a technologist to perform this measurement. as a global leader in standardization.S. and different consensus timelines are all obstacles to harmonization. Trueness is used in this document when referring to the closeness of the agreement between the average value from a large series of measurements to a true value of a measurand.1 This document provides practical guidelines that assist the technologist in counting reticulocytes using flow cytometers and automated blood counters with flow cytometric measurement principles.

that the fundamental meanings of the terms are identical in many cases. flow cytometry. and revised appropriately during the next scheduled revision of this document. reticulocyte. reticulocyte hemoglobin content viii . new methylene blue (NMB). Key Words Anemia diagnosis. All terms and definitions will be reviewed for consistency with international use. red cell. immature reticulocyte fraction. the terms are defined along with explanatory notes in the guideline’s Definitions section. erythropoiesis.Number 8 NCCLS Users of H44-A2 should understand. erythrocyte. reference method (reference measurement procedure). however. and to facilitate understanding.

Definitions) is the new methylene blue (NMB) staining of blood collected in ethylenediaminetetraacetic acid (EDTA) anticoagulant. Approved Guideline—Second Edition 1 Scope This document outlines the essential factors that affect automated reticulocyte counting using flow cytometric principles.1 Interlaboratory studies of flow cytometric methods have shown some intermethod bias in reticulocyte counts. The designated Class C comparative method (see Section 4. Although primarily directed at automated or flow cytometric reticulocyte counting. Standard precaution and universal precaution guidelines are available from the U. refer to the most current edition of NCCLS document M29—Protection of Laboratory Workers from Occupationally Acquired Infections. All rights reserved. Standard precautions cover the transmission of any pathogen and thus are more comprehensive than universal precautions which are intended to apply only to transmission of bloodborne pathogens. 1996. several flow cytometric and image-analysis methods are now available.2 The production of more precise and true reticulocyte data is necessary before these data can be more generally applied to improve patient care. Methods to ensure acceptable precision and trueness of the calibration and quality control of the automated reticulocyte counting methods are outlined. portions of this document are also applicable for the manual/visual and image analysis methods for reticulocyte enumeration. 2 Introduction Reticulocyte counts have been used for many years to estimate the erythropoietic activity of the bone marrow.36[suppl 2S]2S-18S). and (MMWR 1988. Flow Cytometry.” Standard precautions are guidelines that combine the major features of “universal precautions and body substance isolation” practices. as well as maturation estimates.37:377-382. 1 .1:53-80). all human blood specimens are to be treated as infectious and handled according to “standard precautions. and Supravital Dyes). Additionally. Centers for Disease Control and Prevention (Guideline for Isolation Precautions in Hospitals. (MMWR 1987. generated by flow cytometers employing different reagents and instrumentation. 387-388). These include methods to calibrate.Volume 24 H44-A2 Methods for Reticulocyte Counting (Automated Blood Cell Counters. ©NCCLS. and also to control this analysis. Procedures to develop appropriate reference ranges are also outlined. Methods used to express the relative and absolute maturation of reticulocytes are discussed. While in the past this has been done using manual/visual methods. Standard Precautions Because it is often impossible to know what might be infectious. Methods to harmonize the various reticulocyte technologies are provided in this guideline.Vol 17. CDC. For specific precautions for preventing the laboratory transmission of blood-borne infection from laboratory instruments and materials and for recommendations for the management of blood-borne exposure. primarily in an effort to guide in the standardization of these additional clinically useful reticulocyte parameters. termed “immature reticulocyte fraction (IRF)” which is available on many of the automated blood counters using multiparameter flow cytometric principles. An NCCLS global consensus guideline.S. this document addresses principles and quality control of the automated reticulocyte parameters related to reticulocyte maturation. Infection Control and Hospital Epidemiology.

under specified conditions. NOTES: a) The clinical disorder must be defined by criteria independent of the test under consideration. NOTE: The cumulative examination pathway resembles the battlement of a castle. fluorochromes (e.g. by a measurement station.g. 2 An NCCLS global consensus guideline. mean ± SD) ascertained during preparation and confirmed in use. a positive result and identification of the patients who have a disease). thiazole orange). anti-CD71) are counted in a flow cytometer or automated blood cell counter that was specially designed. for this procedure. intended for use in the quality control process.. the relationship between values of quantities indicated by a measuring instrument or measuring system. as the cell passes through the measuring station. and the corresponding values realized by standards (VIM93)..5 Carry-over – The discrete amount of analyte carried by the measuring system from one sample reaction into subsequent sample reactions. ©NCCLS. c) See the Note on Terminology in the Foreword. or lyophilized preparation. Fluorochrome – A chemical compound that has the property of absorbing light at one wavelength and emitting light of a longer wavelength.1217). kit.g. solution. thereby erroneously affecting the apparent amounts in subsequent samples. All rights reserved. or artificially derived material. or test system. b) The expected reaction or concentration of analytes of interest are known within limits (e. to provide a known relationship between the measurement response and the value of the substance being measured by the test procedure (42CFR493. or changes in electrical impedance.S. or monoclonal antibodies (e. or otherwise modified or adjusted. NMB). Flow cytometry – A methodologically oriented subdiscipline of analytical cytology that measures cells in suspension in a liquid vehicle as they pass. c) Control materials are generally not to be used for calibration in the same process in which they are used as controls. Control material – A device.4 Calibration – set of operations that establish. or values represented by a material measure or a reference material. typically one cell at a time. Diagnostic sensitivity – The proportion of patients with a well-defined clinical disorder whose test values are positive or exceed a defined decision limit (i. NOTE: See the definition of Measurand.g. . light emitted (fluorescence) by the cell.Number 8 NCCLS 3 Principle Reticulocytes that are stained with one of a number of different supravital stains (e. Bias – The difference between the expectation of the test results and an accepted reference value (ISO 3534-1). calibration is the process of testing and adjustment of an instrument.e.. absorbed light. 4 Definitions Accuracy (of measurement) – Closeness of the agreement between the result of a measurement and a true value of the measurand (VIM93)3. Battlement pattern – A method of studying a blood film in which the slide is moved from side to side (or end to end) over acceptable examination areas. b) The term Clinical sensitivity was formerly used in this document. NOTE: The measurement represents transformations of changes in the output of a detector (or detectors) due to changes in scattered light.. NOTES: a) Control material may include a pool of collected human or animal specimen. Code of Federal regulations. below..3 NOTE: According to the U.

NOTES: a) This definition is analogous to a Reference measurement procedure in ISO terminology. the range of values An NCCLS global consensus guideline. b) Previously called Reticulocyte maturation index.8 Reportable range – The range of test values over which the relationship between the instrument. “substance” concentration is a quantity that may be related to a particular analyte. 3 .3 NOTE: This term and definition encompass all quantities. Miller disc – An optical micrometer or reticule that is placed in the optical light path of a microscope. kit. For example.. in which exact and clear descriptions of the necessary conditions and procedures are given for the accurate determination of one or more property values. or system's measurement response is shown to be valid.e. 95% of individuals that are presumed to be healthy or normal. the examiner determines the ratio of one cell type (e. Reference interval – The range of test values expected for a designated population of individuals. age-matched individuals.g. while the commonly used term “analyte” refers to a tangible entity subject to measurement. b) Using techniques as described in Section 7. b) For measurements having a Gaussian or log-normal distribution only. NOTES: a) For example. typically a population of healthy.7 Qualified examiner – A person with special training and recognized skills in peripheral blood cell morphology. and in which documented trueness and precision of the method are commensurate with the method’s use for assessing the trueness of other methods for measuring the same property values. and who is qualified according to the criteria detailed in NCCLS document H20—Reference Leukocyte Differential Count (Proportional) and Evaluation of Instrumental Methods. Precision – The closeness of agreement between independent test results obtained under stipulated conditions.g. NOTES: a) This has been expressed in terms of mean fluorescence intensity using thiazole orange on multiparameter flow cytometry instruments and as a fractional expression of the subpopulation of the reticulocytes having the highest fluorescence RNA intensity or RNA content 6 . According to ISO. JW Miller of the National Institutes of Health (NIH). especially in assessing the trueness of other measurement procedures for the same quantity and in characterizing reference materials (adopted from ISO/DIS 15193. Linearity – The ability (within a given range) to provide results that are directly proportional to the concentration [amount] of the analyte in the test sample. All rights reserved. or for assigning reference method values to reference material. a reference measurement procedure is a thoroughly investigated measurement procedure shown to have an uncertainty of measurement commensurate with the intended use. ©NCCLS. Measurand – particular quantity subject to measurement (VIM93). mature erythrocytes). the final analytical answer) rather than the raw instrument output.Volume 24 H44-A2 Immature reticulocyte fraction – A quantitative expression of the maturation state of the entire reticulocyte population in the peripheral blood. Reference method – A thoroughly investigated method. NOTE: It is expressed numerically as the Standard deviation or Coefficient of variation of the results in a set of replicate measurements. NOTE: Linearity typically refers to overall system response (i. NOTE: Precision is not typically represented as a numerical value but is expressed quantitatively in terms of imprecision—the standard deviation (SD) or the coefficient of variation (CV) of the results in a set of replicate measurements (ISO 3534-1). NOTES: a) It is designed with a large square that contains a smaller square exactly one-ninth the area of the large square (see Figure 1A).. c) The disc was invented by Dr. NOTES: a) For this document.. reticulocytes) to another cell type (e. this may be represented by the population mean plus and minus two standard deviations.3. Imprecision – Dispersion of independent results of measurements obtained under specified conditions.

that is. and color slope with multicolor capability. Static cytometry – A recently introduced technology which differs from flow cytometry in that fluorescently stained cells are optically measured in a fixed-volume capillary. i. Specimen (patient) – The discrete portion of a body fluid or tissue taken for examination. b) See Measurand above. above.g. Sensitivity – The change in response of a measuring system or instrument divided by the corresponding change in the stimulus (modified from VIM93-5. Bias.e. contain hemoglobin as the primary oxygen-carrying molecule. above.4 NOTE: Trueness is usually expressed numerically by the statistical measure bias that is inversely related to trueness. Tolerance limits – Specified limits for allowable error.10). Specificity – The ability of a measurement procedure to measure solely the measurand. or analysis of one or more quantities or characteristics to determine the character of the whole. 2) An erythrocyte that. the cell must contain two or more distinct blue-staining granules (NMB-reticulocyte) that are visible without requiring fine focus microscope adjustment on the individual cell to confirm their presence.. contains stainable nucleic acids (i. Reticulocyte (retic) – 1) Transitional red cells between nucleated red cells and the so-called mature erythrocyte.. NOTES: a) To be identified as a reticulocyte by visual microscopy. or other sources are within defined limits. 4 An NCCLS global consensus guideline. study. b) The granules should be away from the cell margin to avoid confusion with Heinz bodies. cellular RNA) or when stained with fluorescent reagents with specificity of binding to nucleic acids shows increased cellular fluorescence. NOTE: Measurement of the hemoglobin content can be performed and quantitated in terms of mean picogram of hemoglobin per cell.3 NOTES: a) The sensitivity may depend on the value of the stimulus. ©NCCLS. NMB). Sample (patient) – A sample taken from the patient specimen and used to obtain information by means of a specific laboratory test. Trueness (of measurement) – The closeness of agreement between the average value obtained from a large series of test results and an accepted reference value. See also Accuracy. cell size. .3 See Measurand defined..e. b) Gating strategies are similar to flow cytometry and include cell staining intensity.10 NOTE: a) Specificity has no numerical value in this context. (ISO 3534-1).9 Reticulocyte hemoglobin content (CHr) – Reticulocytes.11 Supravital dye – A dye used to stain living cells already removed from an individual. NOTE: It can lack some of the functional characteristics of blood. imprecision. where errors due to nonlinearity. a set of values of measurands for which the error of a measuring instrument is intended to lie within specified limits.Number 8 NCCLS (in units appropriate for the analyte) over which the acceptability criteria for the method have been met.3 b) The sensitivity depends on the imprecision of the measurements of the sample. when stained with a supravital dye (e. similar to the mature erythrocyte. Stabilized blood product – A material prepared from blood cells treated to prolong its usefulness as a quality control material. c) This is similar to the VIM definition for Measuring range or Working range. b) The reportable range of the assay should be established prior to beginning the clinical evaluation. Also see the Note on Terminology in the Foreword. NOTE: Limits should depend on both the effect of the error on the clinical significance of a test and on what is technically achievable. and the image of the cells is plotted and counted providing an absolute cell number in a defined volume. All rights reserved. NOTE: a) In this technique the laser continuously scans down the length of a disposable capillary.

2 Skin Puncture Blood Collection Please refer to the most current edition of NCCLS document H4—Procedures and Devices for the Collection of Diagnostic Blood Specimens by Skin Puncture for recommendation on the collection of diagnostic blood specimens by skin puncture. EDTA is available as di/tripotassium and disodium salts.Volume 24 H44-A2 5 5. 5. a unique patient identification number.1 Specimen Collection and Storage Patient Information A test requisition should accompany all specimens and must include patient demographic information.2. such as heparin or citrate. When multiple specimens from the same patient are collected for analysis. it should state the source of the specimen. and should ideally state the reason why the test is requisitioned. For additional information. When shipped.7 to 5. see the most current version of NCCLS document H18— Procedures for the Handling and Processing of Blood Specimens.4 Labeling of the Specimen The specimen should be labeled with a unique patient identifier as on the requisition form. The date and time of collection of every specimen should be noted. 5. ©NCCLS. Biosafety Level 2 (BSL-2) practices should be followed during the collection and handling of body fluids. it should be noted that mean cell volume (MCV) changes over time and increases as a result of lack of oxygen and other metabolic processes.2 Specimen Collection Techniques As a rule all body fluids. and standard precautions must be followed. 5. Specific labeling of specimens as infectious is not required. such as umbilical cord aspiration or amniotic fluid. care should be taken that shipment temperatures do not exceed room temperature. have not been thoroughly evaluated for reticulocyte counting. bone marrow.2. Pertinent laboratory information should be made available when appropriate and possible. 5 . Other anticoagulants.2. An NCCLS global consensus guideline. either using a flow cytometer or a hematology analyzer. because all specimens should be regarded as potentially infectious. An accurate red cell count should be determined on the same sample being used for the reticulocyte count. should be considered potentially infectious. or other aspirates.4 µmol/mL or 1. All rights reserved.2.5 to 2.2.5 Specimen Handling and Packaging Please refer to the most current government regulations for detailed information on specimen handling and packaging.2 mg/mL) is the acceptable specimen. whether they be blood. 5. If measurement of red cell volume is part of the assay. If the specimen is other than venous blood.1 Venipuncture Please refer to the most current edition of NCCLS document H3—Procedures for the Collection of Diagnostic Blood Specimens by Venipuncture for detailed information on the collection of blood specimens by venipuncture. NCCLS document H1—Evacuated Tubes and Additives for Blood Specimen Collection gives details about EDTA as an anticoagulant.3 Anticoagulants Whole blood collected into EDTA (final concentration 3. 5. 5. the specimens should have a unique specimen identifier or hazard label.

storage conditions (time and temperature). 5. All rights reserved. ©NCCLS.. the small volume sample may not affect final results.4. but severely clotted samples should be rejected. . however. 6 An NCCLS global consensus guideline.3 Partial Draw When the blood specimen is severely underdrawn into a blood collection device. Samples exhibiting some evidence of hemolysis may still be tested by the laboratory.1 Hemolysis Hemolysis indicates that the blood has been exposed to conditions that can cause erythrocyte lysis. analysis. the reported results should note that the presence of hemolysis may present a potential complication. Reported results should note evidence of exposure to extreme temperatures as a potentially complicating issue. however. Samples exhibiting evidence of blood clots may still be tested by the laboratory. If no hemolysis has occurred. however. and sample preparation procedures.e. Samples designated for overnight shipment should be packaged on wet ice or similar material to ensure near-refrigerated temperatures without the danger of sample freezing. and interpretation. Sample integrity is a composite product of the anticoagulant.4. the results should note that the presence of a clotted sample may present a potential complication or may not reflect the true reticulocyte level in the patient.4. refrigerated temperatures (4 °C) should be preferred to prevent deterioration of the sample. exposure should be noted for consideration during preparation. Loss of specimen integrity is judged by the lysis or loss of red cells. there should be a suspicion that the sample has been exposed to extremely hot or cold temperatures. Laboratories should establish specific guidelines for sample collection and storage based upon the assay method and anticoagulants employed.4. Samples exhibiting evidence of such exposure to extreme temperatures may still be tested by the laboratory. 5. 5.2 Clotted Blood A small blood clot (not evident upon visual inspection) may not influence the outcome of testing. the hypertonic conditions by the anticoagulants present in the tube may be deleterious to the cells.4 Sample Evaluation Sample evaluation is necessary to ensure sample integrity as well as proper sample handling and mixing prior to analysis. but for long-term storage (greater than six hours). 5. If a sample arrives in the laboratory after prolonged storage (i. 5. These are general guidelines based upon the collective experience of the subcommittee.Number 8 NCCLS 5.5 Improper Specimen Labeling Improperly labeled specimens should be rejected by the laboratory.3 Sample Integrity Handling and transportation procedures must maintain the viability of samples.4. Samples can be maintained at room temperature (18 to 22 °C). which may be determined by cell counting or the increase in free hemoglobin level in the plasma component of the specimen. Visibly hemolyzed samples should be rejected. but should be noted as a potential complication. greater than six hours after blood collection) in a frozen state or at ambient temperature.4 Extreme Temperatures If the sample was mailed to the laboratory. it may have been exposed to extreme temperatures. 5.

Equal volumes of NMB stain and blood are mixed in a glass or plastic test tube. reproducibility. There should be no visible hemolysis.1 gram of NMB crystals to 100 mL of iso-osmotic phosphate buffer. If sample analysis is delayed. that the specimen be stored in such a way that it remains stable until the test is performed. there are inherent limitations in the imprecision level.Volume 24 H44-A2 5. An NCCLS global consensus guideline.5 Specimen Storage It is recommended that the reticulocyte count be performed promptly after the collection of the blood specimen. there is an apparent in vitro maturation and subsequent disappearance of some of the reticulocytes. reticulocyte counting should be done within six hours after blood collection. However. Samples stored at 2 to 6 ºC may be stable for up to 72 hours. This maturation is both time.1 Stain The supravital dye. the proportion of dye to blood should be adjusted accordingly.14 It has some advantages of purity and the dye does not precipitate on blood films. being a manual and subjective method. Store in a brown glass bottle at 2 to 6 ºC.4 (18 mL of 150 mmol/L NaH2PO4 and 82 mL of 150 mmol/L Na2HPO4). ©NCCLS. an appropriate volume must be filtered through filter paper to remove any precipitated dye and any other particulate matter. it is not recommended that any inhibitors of such maturation be added to the sample. However.000 cells and must be done by previously qualified examiners whose ability was verified using the method outlined in NCCLS document H20—Reference Leukocyte Differential Count (Proportional) and Evaluation of Instrumental Methods. It should have a dye content of approximately 90%. At this temperature. this may vary with the counting system being used. pH 7.12 However. NMB (basic blue 24. Initial calibration and/or verification of reticulocyte flow cytometers should be carried out by counting patient samples stained with NMB or another previously calibrated. the sample should be refrigerated. After preparation. Before use.15 6.and temperature-dependent11 and can be expressed as the number of reticulocytes that mature and are therefore not counted after the delay in analysis. a purified component of polychrome methylene blue. and sensitivity of the NMB method. Under such conditions. All rights reserved. 7 . This loss can be up to 20% per 24 hours at room temperature. 6.e. Azure B. In case of anemic or polycythemic blood. has been found to be equally acceptable as reticulocyte stain. the dye solution has a shelf life of about one month. automated reticulocyte counting method and covering the entire range of reticulocyte counts claimed to be accurately enumerated by the instrument.. or alternatively.13. When the original specimen is kept at room temperature. Add 0. the manufacturer’s stability recommendations for their reticulocyte method should be confirmed by the laboratory. the solution should be shaken at intervals over 24 hours. coefficient of variation [CV] ≤ 3%). color index (CI) 52030) is the stain of choice. With most reticulocyte methods. and the red cell count should be stable (i. 6 Correlation Method—New Methylene Blue Visual Microscopy The NMB staining method has been recommended as the comparability method for reticulocytes. The NMB procedures should be done based upon a minimum count of 2. special storage conditions are not required. Presently.2 Slide Preparation A specimen of whole blood collected in EDTA is mixed by gentle inversion ten times or until the complete resuspension of cells is accomplished.

Number 8 NCCLS After staining at ambient temperature for three to ten minutes. and the cells are ideally spaced. and also the right. it must comply with all applicable safety requirements. Table 1 (Section 7. The “edge rule” states that red cells touching the lower.16 If the edge rule is ignored. these examples have canceled out the data from n fields. . (See the Standard Precautions statement after Section 1. to obey the “edge rule” when counting red cells (as well as reticulocytes) in the small square of the Miller disc. The small square may be either in the center or at a corner of the large square..2% = red cells counted = reticulocyte Figure 1A. many more than one field would be counted. particularly with higher reticulocyte counts. a significant bias will be found between the Miller disc count and the count done without the use of the Miller disc (i. side of the square should not be counted. If a Miller disc is employed. The red cells should be closely spaced but not touching or overlapping.4) shows the number of red cells to be counted to obtain the necessary level of precision.17 Three examples of this are given below. All rights reserved. For clarity.e.) 6. If a blood sample has a true reticulocyte count of 1.3 Counting Procedure The blood films are examined under low power (100x magnification) to ensure the uniform distribution of the red cells. 8 An NCCLS global consensus guideline. ©NCCLS. (Please refer to the most current version of NCCLS document H20—Reference Leukocyte Differential Count (Proportional) and Evaluation of Instrumental Methods. In actual practice.2%.) If a slide spinner is used to prepare the films. the calculation of the reticulocyte percentage in Figure 1A will be: Retics (Large Square) Red Cells (Small Squarea) x 9 = = x 100 (1) 1 x 100 9x9 1. If satisfactory. Miller Disc Reticulocyte Count with Ideal Red Cell Distribution a Any reticulocytes in the small square are to be counted as red cells. Films covering about two-thirds of the surface are made on glass microscope slides and allowed to dry rapidly in warm air. the count without the use of a Miller disc is about 30% higher). The total number of red cells to be counted varies with the reticulocyte percentage as well as the desired precision. the stain and blood mixture should be well mixed just before placing a drop of blood on the glass slide. it is important. the film is examined with an oil immersion lens moving from field to field in a battlement pattern.

All rights reserved.Volume 24 H44-A2 If the Miller disc is shifted so that there are many red cells touching the square sides and if the edge rule is ignored (incorrect use of Miller disc): Retics (Large Square) Red Cells (Small Squarea) x 9 = x 100 1 x 100 16 x 9 0. An NCCLS global consensus guideline. ©NCCLS. 9 .7% = = red cells counted = reticulocyte Figure 1B. This result agrees with the correct result in Figure 1A. Incorrect Use of Miller Disc If the same field as in Figure 1B is counted and the edge rule is imposed (correct use of Miller disc): Retics (Large Square) x 100 Red Cells (Small Square.b including two edges only) x 9 = 1 x 100 9x9 1.2%c = b c Any reticulocytes in the small square are to be counted as red cells.

it is recommended that such counting methods not be employed for calibration or linearity verification of automated reticulocyte counting methods unless a minimum of 1. from the reticulocyte analyzer itself.000 red cells has been counted. ©NCCLS. Retic Count (x109/L) = (4) RBC (x1012/L) x Retics (%) x 10-2 Due to the above-described potential for counting errors and the potential for counting bias using the Miller disc. . Abs. multiply the percent (ratio) of reticulocytes by the red cell count determined on the same blood sample using an adequately calibrated and controlled hematology instrument or. the reticulocyte percentage is calculated as follows: Retics (%) = Total Number of Retics in 20 large squares x 100 Total Numbers of RBCs in 20 small squares x 9 (3) To determine the absolute number of reticulocytes. because the area of the small square equals one-ninth that of the large square. The percentage of reticulocytes is calculated as follows: Retics (%) = (2) Total Number of Reticulocytes x 100 Total Number of Red Cells + Retics If a Miller disc is used. All rights reserved. Correct Use of Miller Disc 6.Number 8 NCCLS = red cells not counted = red cells counted = reticulocyte Figure 1C. if available.4 Calculations If a Miller disc is not used. 10 An NCCLS global consensus guideline. the number of reticulocytes and the number of red cells (including reticulocytes) are tallied separately.

which makes possible the discrimination of reticulocytes from mature red cells. erythroblasts. 7. All rights reserved.3 Method(s) of Operation The methods of operation must be clearly described in the operator’s manual.e. malaria infection. i.. discrimination is achieved by means of the staining of reticulocyte RNA with fluorochromes or supravital dyes. Binding of fluorochromes to RNA makes the reticulocytes fluorescent. as these parameters are early indicators of a normal or abnormal erythropoietic regeneration). makes the reticulocytes capable of absorbing more light than unstained red cells. The fluorescence of stained reticulocytes should be high enough to permit the discrimination from autofluorescent red cells. This includes samples with near-zero reticulocytes (such as after myeloablative chemotherapy) as well as samples with very high reticulocyte counts (such as in hemolytic anemias. 7. anthracyclines) or pathological conditions (e. such as leukocytes. strict adherence to such recommendations is mandatory. ©NCCLS. even in the case of fully automated methods. Binding of supravital stains.. erythroblasts.2 Sample Preparation for Flow Cytometry Blood samples are prepared according to the methods recommended by the instrument manufacturer. and methods and warnings used to identify potentially interfering substances or cellular elements. 11 . Discrimination from other cells present in peripheral blood. mature red cells have a low level of autofluorescence due to the porphyrin ring in the hemoglobin molecule.. recommended methods for calibration and quality control of all reportable reticulocyte parameters. the proportion of immature reticulocytes should be accurately measured in the same ranges (especially at the lowest reticulocyte counts. All dyes should be studied for interfering conditions. 7. Similarly. and platelets. To obtain acceptable reticulocyte counts.1 Reportable Range (Measuring Range) Reticulocyte counters should provide precise and true reticulocyte measurements over the entire reportable range (measuring range).Volume 24 H44-A2 7 Specifications for Automated Blood Counters and Flow Cytometers to Be Used for Reticulocyte Counting Automated reticulocyte counting systems based upon flow cytometric principles or static cytometry methods used for reticulocyte counting should measure a specific cellular property.g. the system used to discriminate unstained red cells from stained reticulocytes. such as commonly used medications that might quench signals (e. An NCCLS global consensus guideline. and platelets. the immature reticulocyte fraction (IRF). Heinz bodies. or following vitamin B12 treatment of megaloblastic anemia). the system used to quantify immature reticulocytes. such as new methylene blue or oxazine 750. At the present time. such as leukocytes. is also necessary.g. or Howell-Jolly bodies). in association with measurements of size and internal structure. The following points require detailed explanation: • • • • • the system used to separate the relevant cell populations (red cells plus reticulocytes) from irrelevant cell populations. The same measurements are used for the differentiation of mature reticulocytes with average RNA content from immature reticulocytes with increased RNA. because it is such samples where the clinical utility of these measurements is maximal.

electronic gains. Instrument calibration is the adjustment of an analytic process under specified conditions to obtain accurate measurement results. ©NCCLS. All rights reserved. Optical alignment and signal amplification settings can require the use of artificial particles of standardized size and density. lesser precision (CV ~5%) is probably adequate. Table 1. Each reference analysis should count a minimum of 2. such as adjustment of photomultiplier voltages. but the level of imprecision is not the same and likely higher than that achieved by counting ninefold the cells without the ocular assist device. The approximate numbers of red cells required to be counted for levels of imprecision of 2. In the event that disagreement beyond predictable limits occurs between the two analysts.000 2. covering the reportable range (measuring range) of the instrument’s reticulocyte counts. Results. These films should be single-blind coded by the coordinator for distribution to the participating prequalified examiners.20 0. since the small square in which the red cells are actually counted is 1/9 the size of the larger square. as well as of fluorescent microbeads. and 10% at various reticulocyte proportions are given in Table 1. semiannual or following major instrument service work) assessment of the linearity of reticulocyte counting methods should be performed to verify the trueness of the reportable range (measuring range) of the automated instrument.900 900 400 100 Retic Percentage 0. Calibration should be performed in accordance with the method specified by the manufacturer. should be correctly set and periodically checked according to the manufacturer’s specifications as a part of the internal quality control procedure.01 0.e.500 22.900 1. the actual RBCs counted would be 1/9 of these numbers. for routine quality control purposes. retain the result. a more precise reticulocyte count (CV ~2%) is required. Three NMB-stained blood films are prepared on each sample. identified by code number.000 red cells or up to the required number as noted in Table 1. fluorescence. two for reticulocyte counting by prequalified examiners.900 4. If the arbitrator’s count falls between the other two.600 7. Even for a reference method (reference measurement procedure) the table shows that acceptable precision cannot be readily achieved when the reticulocyte count is in the range of 1 to 2%. the arbitrator should perform one additional reticulocyte count for verification of the results.500 10. otherwise.10 0.50 * 2% 247. Technical aspects. should be submitted to the coordinator for evaluation. If a Miller disc is used. the third for use in arbitration of the counts. should it be required.500 122. 12 An NCCLS global consensus guideline. 5.4 Instrument Settings and Calibration Optical and electronic components of the instrument should be set according to the manufacturer’s directions.02 0. and absorption signals.600 1.500 47. use at least three fresh blood samples for the calibration. The coordinator should determine the mathematical mean value from duplicate reticulocyte counts done by a minimum of two prequalified examiners. laser output.500 These RBC numbers are the total RBCs to be counted if a Miller disc is not being used. RBC Number to Be Counted for Required Precision* Coefficient of Variation 5% 10% 39. If on-site instrument calibration is required..600 19.600 3. Instrument settings should provide the maximum intensity with minimum variability of scatter. and signal to noise discrimination.05 0. For initial calibration. Regular (i.600 400 9. .Number 8 NCCLS 7.

malaria or other intracellular parasites. It is recommended that a minimum of 26 blood samples be examined. such as high levels of Howell-Jolly bodies. the means of duplicate analyses by the comparability method are compared with mean values of duplicate analyses by the test (instrument) method. 13 . Instrument calibration should be done using the manufacturer’s specifications at the time of initial installation. thrombocytosis. 8. and sources of interference. The assessment should include the direct flow cytometric measurement. lymphocytosis.19 However. The evaluation should also include samples with abnormal cells or other blood components that can interfere with reticulocyte counts.1. so decreased and increased erythropoietic activity in patients can be identified. is intended for the initial examination of device accuracy by determination of agreement with the comparability method. Comparability has been defined as the ability of the instrument to produce results that agree satisfactorily with those obtained by routine procedures. including neonatal samples. as yet. or agreement. All calibration and recalibration procedures and adjustments. is even more problematic due to the lack of commercial calibrators or standards. Because there is. which usually is the proportion of stained cells relative to the total red cell population. 8. as indicated by the instrument manufacturer or as established by the laboratory.1 Comparability (Correlation) Method Trueness can be established by the acceptance of values obtained with another previously calibrated method (see Section 6) and comparing the test instrument to these “reference” values. All the analytical aspects can be evaluated using the ICSH guidelines for the evaluation of blood cell analyzers18 using samples at all the levels of reticulocyte counts that can occur in clinical practice. Reticulocyte percent values should be converted to reticulocyte count number by use of an RBC count done on a well-calibrated and controlled cell counter.Volume 24 H44-A2 take the mean of the two closest counts. should be documented completely with written records. ©NCCLS. All rights reserved. it is more appropriate to refer to this as a method of comparability. trueness.1 Diagnostic Sensitivity Studies Sample selection. linearity. or whenever quality control studies or proficiency testing indicates a potential problem with instrument performance. whenever major preventive maintenance or critical parts replacement is done that can influence test performance. or based on routine laboratory results that adequately represent the entire claimed performance range. disease state. as well as the proportion of immature reticulocytes relative to the total reticulocyte population. Samples collected for use in the abbreviated accuracy validation procedure may be selected randomly from routine samples. Calibration of reticulocyte-related parameters. The variability in IRF among the various reticulocyte staining methods is due to difference in both reagent and data analysis routines. or validation of accuracy and linearity. as noted in this section. should be done periodically. Additional samples should be selected on the basis of clinical history. correlation. no consensus value for “truth” (including the NMB method). such as the immature reticulocyte fraction (IRF). 8 Performance Testing of Automated Blood Counters and Flow Cytometers to Be Used for Reticulocyte Counting Performance tests for automated instruments for reticulocyte measurements should include assessment of precision. or erythroblastosis. and all calibration verification or validation. sample aging. as a guiding principle manufacturers should seek a calibration that allows the reference range on normal adult samples to detect low and high IRF values. carry-over. comparability. Recalibration. with approximately 50% representative of normal range samples and the remainder representative of abnormal range samples (25% decreased and 25% increased An NCCLS global consensus guideline.19 For reticulocyte counting.

The formula for short-term standard deviation (SDs): SDS = ∑d 1 n 2 i 2n (5) where: n = number of samples. Insert these differences (di) into formula (4). and not having donated blood within the last two months. 8. Calculate the short-term imprecision at the three levels of reticulocyte counting using formula (4): Convert SD to Coefficient of Variation (CV) using Equation (6): CV (%) = SD Xa x 100 (6) Note that Xa is the mean of all values. Elevated reticulocyte count samples can be obtained from patients with hemolytic anemias or reactive erythrocytosis. not the mean of the differences. it is important to complete the precision study (see Section 8. newborns. For example. and di = difference between duplicates for sample i. NCCLS Pediatric reference ranges are higher than adults. Depressed reticulocyte count samples can be found using samples from patients with aplastic anemia or postcytotoxic chemotherapy. A minimum of ten samples in each tertile will provide a useful estimate of standard deviation (SD). leukocyte count. platelet count). All samples used in this accuracy validation protocol should be stored according to instructions provided in the section on patient specimen collection and handling (see Section 5). All rights reserved. To utilize this method effectively.1. .1). and make sure to divide the data into the tertiles according to the recommended ranges for reticulocyte samples (Section 8. because accuracy studies can be invalidated by a failure 14 An NCCLS global consensus guideline. ©NCCLS. 8. Abnormal samples should be selected from patients with known or suspected hematological conditions that would have either elevated or depressed reticulocyte counts.3 Procedure Before beginning the device accuracy validation procedure.2 Calculation of Short-Term Imprecision Determine the absolute differences between the duplicate determinations.2) with acceptance of test results.Number 8 reticulocyte counts). hemoglobin and indices. establishment of an acceptable limit for duplicate analysis replication for a 200 x 109/L reticulocyte count might be unacceptably wide for a reticulocyte count of 40 x 109/L.1.1. being free of recent illnesses. Samples should represent as wide as possible a range of reportable values.1. each laboratory should establish acceptable limits of reproducibility throughout the range of samples that will be tested. Higher values of n will give more robust estimates of SD. particularly in infants and The reference or normal samples should be selected by qualification of the donor as having no detectable hematological abnormality (red cell count.

pFC) nR and nFC are the total number of red cells and reticulocytes counted.4. using both methods within six hours of the time of specimen collection. particularly during comparison of IRF parameters. Each sample should be analyzed in duplicate.0000027 + 0. This method may also be used for the periodic validation of instrument performance by testing a limited number of samples. The example uses a theoretical model based on an instrument counting 30. The automated reticulocyte counting method should follow the device manufacturer’s recommended analytic procedures. This procedure requires careful examination of the same samples by both the comparative method and the automated method being validated.1.000 cells.000 cells and the reference method (reference measurement procedure) counting 4.1 Accuracy Schema for Data Analysis for Accuracy Because there are two possibilities for error.00000033 = 0. It is advisable that the laboratory establish a scheme for analysis of the samples by testing. Concurrent use of these data for definition of instrument imprecision is noted in Section 8. 8. SEp = VarR + VarFC (9) Combining the variances of the test and reference methods (reference measurement procedures) above: SEp = 0. Var R = 0.99 = 0. All rights reserved. The reticulocyte value ÷ 100 is 1%. while determining the mathematical mean for comparison with the reference method (reference measurement procedure).1.00000033 30000 A combined variance of reference and flow cytometer methods is calculated by adding individual variances together given that the reticulocyte percents are obtained independently by each method.pR) and qFC = (1.00000247 4000 VarFC = 0.Volume 24 H44-A2 of the precision test procedures.2.01 x 0.99 = 0. reference and test. Example: An example of this calculation is provided below. ©NCCLS.01 x 0. the variances for each of them are calculated. • Calculation of the variance of the reference method (reference measurement procedure) (VarR): Var R = (p R x q R ) nR (7) • Calculation of the variance of the flow cytometer (VarFC): Var FC = (p FC x q FC ) n FC (8) where. 15 .00167332 An NCCLS global consensus guideline. The square root of this sum represents the combined standard error of test and reference.4 8. pR and pFC are reticulocyte percents qR = (1.1.

Number of Samples z-value for simultaneous 95% confidence intervals 26 30 40 50 60 70 80 90 100 150 200 3. The 95% confidence limits for a reticulocyte percent = 1% are.00167332 x 1.23 3.29 3. the simultaneous margin of error for all comparisons is: E = ± SEp x zα/(2n) .10 3.48 3.01 ± E = 0. if the reticulocyte percent obtained by the reference methods (reference measurement procedures) is 1% then the test method can be as low as 0.96 = 0. .000 cells. All rights reserved. 0.00327 Lower limit = 0. Where zα/(2n) = 3. The margin of error for comparing test and reference methods (reference measurement procedures) is calculated as.14 3.38 3. and 26 blood samples is given in Table 2. E = ± SEp x zα/2 .42 3. (10) where zα/2 = 1.67% Upper limit = 0.01 ± 0.33%.96 for the 95% confidence level.Number 8 NCCLS The reticulocyte count is a binomial distributed variable that can be approximately distributed as normal when both quantities pR x nR and qR x nR are greater than 5.0133 or 1. This requirement is generally satisfied since the total number of red cells and reticulocytes counted is relatively large. ©NCCLS.10 for the simultaneous 95% confidence level and n=26. Z-values for the simultaneous 95% confidence intervals and different number of samples are given in the table below.0067 or 0.000 cells. then the assumption of accuracy fails for that particular reticulocyte level.67% and as high as 1. 16 An NCCLS global consensus guideline. instrumental counts of 30.45 3. When a number of n blood samples are used for comparing two methods.33% Thus.01 ± 0.34 3.59 3.66 An array of reticulocyte values with reference counts of 4. If the value obtained by the test method is outside these limits.

00429 8 0.00001410 0. or X-axis) represents the mean of duplicate reference observations.080 0. then the data should be examined for other sources of error.00053 1 0.020 0.960 0.1.00001628 0. To verify the calibration of the instrument with a laboratory reference method (reference measurement procedure).1 0.0027 0.0273 0.050 0.0502 0. regression analysis can be used.010 0.00367 6 0.1.00000300 0.00001840 0.00000960 0.0843 Upper Limit 0.2 Graphic Display H44-A2 Lower Limit -0. All rights reserved.930 0.940 0.950 0.900 0.0724 0.4.0751 0.0211 0.990 0. to chance alone).0048 0. Superimposed on this graph is the envelope representing 95% confidence limits of the binomial.910 0.0298 0..0567 0.970 0. or Y-axis) represents the mean of results obtained from duplicate analyses on the flow cytometer (see Figure 2).920 0.00457 9 0.00330 5 0.00000245 0. Accuracy Estimation: Paired Sample Method* Variance Variance Retic % p SEp q nFC =30.00505 * Lower limit values that are negative can be replaced with zero.00000490 0. If more than 5% of the values fall outside the binomial confidence bands.00000065 0.00400 7 0.0152 0.0476 0.e.0942 0.0833 0.0658 0.060 0. 8.090 0.00000033 0.00000128 0. 10 9 8 7 Test (%) 6 5 4 3 2 1 0 0 2 4 Reference (%) 6 8 10 Figure 2.030 0.0614 0.001 0.00287 4 0.00167 2 0.00482 10 0.0389 0.0386 0. ©NCCLS.00000248 0.3 Interpretation of Graphic Display Assuming that the only source of error is due to binomial counting error (i.00000217 0. Comparison of Instrument Performance with Reference Method (Reference Measurement Procedure) 8.1157 A graph of the data in Table 2 is prepared where the abscissa (the horizontal.4. no more than 5% of the values should fall outside of the binomial confidence bands.00000273 0. Calculation of 95% confidence intervals of the predicted values from the regression line can be used to estimate bias.999 0.00002250 0.00000097 0.0007 0.Volume 24 Table 2.0127 0. An NCCLS global consensus guideline.980 0.00000158 0.00236 3 0.000 0.000 nR =4.00000188 0.040 0.00000025 0.100 0. the ordinate (the vertical.00002048 0.00001188 0.00000003 0.1049 0. 17 .070 0.00000728 0.

If a sample contains 1% reticulocytes. this same 1% reticulocyte count sample were analyzed on a flow cytometer that counts 30.2 Internal Consistency Testing Because automated reticulocyte counting methods enumerate 30 (or more) times as many cells as the usual manual methods.9 to 1. x = reference count. it is presumed that ten reticulocytes would be counted. x = reference count.4 to 1. ©NCCLS. The confidence intervals for the true reticulocyte proportion for various sample sizes are calculated as follows: P ± z α/2 (p x q) n (14) The usual manual reticulocyte count differentiates 1. di = xi .000 events (cell counts). 300 reticulocytes are counted.(∑ x ) ][n ∑ y . (b) Linear Regression Statistic: y = mx + b where y = instrument count.1%. All rights reserved. 18 An NCCLS global consensus guideline.(∑ y ) ] 2 2 2 (11) where r = correlation coefficient. 8. .000 red cells. reproducibility of the flow cytometric method is expected to be better.Number 8 NCCLS Reference should be made to any standard statistical textbook for the following additional statistical procedures: (a) Correlation Coefficient (r): r= n ∑ xy . m = slope. n-1. and y = instrument count. Paired t-Test Statistic: Calculate. If. The expected performance for this example would be 0. and b = y-intercept.6%. however. n = number of comparisons.yi for each sample d= (12) (c) ∑d i =1 n i n ∑ (d SD = i =1 n i − d)2 n −1 d t= SD n (13) The difference between methods is statistically significant if |t| > tα/2.(∑ x )(∑ y) [n ∑ x 2 . The 95% confidence limits for this reticulocyte count would be 0.

and after routine maintenance. by use of flags triggered by the failure of appropriate gate settings. the most common problems are due to platelet clumps. Analytic procedures for use of these devices should alert the user to the occurrence of these interferents. and intracellular parasites. and previous performance history. precision be examined at three reticulocyte count values. Specific cellular and noncellular elements that can adversely affect the method specificity are included in Table 3. which can stain with the fluorescent dyes used in some reticulocyte methods. such as malaria. The counts of these samples can be approximated by preliminary analysis with the automated method. Instrument manufacturers should clearly indicate any potential interferents in their device support material.3 Specificity of Testing Specificity of reticulocyte analysis can be impaired by other cellular material containing the nucleic acids RNA and/or DNA. which separate cell populations. Table 3. thus. For example. Any RNA. normal (~60 x 109/L).2. All rights reserved. 8. low (≤10 x 109/L).2 Procedure Count each sample of the study samples in duplicate. leukocyte fragments (e. Interferents typically affect the placement of discriminators or gates. 8.. nucleated red cells.2. Of these potential interferents. after service that could affect hydraulic or pneumatic operation.g. A schedule should be established by each laboratory for precision validation depending on frequency of instrument use. Falsely depressed reticulocyte counts can be observed with the presence of cold agglutinins if the analyzer concurrently performs a red cell count.Volume 24 H44-A2 The status of instrument precision should be verified at the time of initial installation. be sure to do two entirely separate procedures starting with the anticoagulated blood sample. and make sure each determination is an independent test..1 Sample Selection It is suggested that. anthracyclines or fluorescein for fundus angiography) Abnormal red cells Paraproteins Cold agglutinins Platelet/erythrocyte coincidence Hemolysis 19 An NCCLS global consensus guideline. 8. if preliminary sample preparation is required. Howell-Jolly bodycontaining red cells. and high (≥200 x 109/L) reticulocyte count samples. aged blood samples or leukemic samples). babesia) Miscellaneous Autofluorescence with porphyria and some drugs (e.or DNA-containing elements within the cells can be stained by the dyes/fluorochromes and.g. Five samples are selected in each range. Known or Potential Interferents Cellular Elements Platelet clumps Basophilic stippling Giant platelets Leukocytes and leukocyte fragments Nucleated erythrocytes Cellular Inclusions Howell-Jolly bodies Heinz bodies Pappenheimer bodies Parasites (malaria. . work load. whenever possible. for the initial instrument performance validation. ©NCCLS. be enumerated by the fluorescencedetection methods of the flow cytometer potentially as reticulocytes.

and because it is unlikely that the nature of the relationship between the test and comparability methods will be known with any certainty below 1. the slope and intercept for regression of the test system and reference method (reference measurement procedure) data may not be 1 and 0. Wash with physiological buffered saline (PBS). (Please refer to the most current version of NCCLS document EP6—Evaluation of the Linearity of Quantitative Measurement Procedures: A Statistical Approach for additional information. Sensitivity must be tested at the limit of clinical performance requirements.00 mL of dilute Sample H into tubes 0. 90.75.1. Precision is determined using the methods and equations in Section 8. the test system should be tested for linearity and sensitivity independent of the comparative method chosen. 0. A good source of low reticulocyte blood is the post-treatment blood from bone marrow transplant patients that will be assayed for platelet and granulocytes. 0. Because the recommended performance level for automated reticulocyte counting includes samples between zero and 1.10. 50. and then add 0. reticulocyte count (~150 x 109/L) and label it “Sample H.90. sensitivity is also related to linearity. centrifuge.0% reticulocytes due to limitations of the comparability method. at the nadir of bone marrow activity. This value should be obtained from the accuracy analysis.1% NMB reticulocytes. to detect reticulocytes between the 0. carry-over. et al 20 give examples of the use of this procedure in reticulocyte counting. 75. Pipette 0. or moderately elevated. and it would then be routinely tested so that the instrument performed in the range necessary for the laboratory. 25.25. 25. Do not use samples with cold agglutinins.75. This might only become known after having analyzed samples falling below the known sensitivity of the instrument. This is a complex issue.01 and 0.1%.) To test linearity and investigate sensitivity. 0. Set up seven tubes labeled 0. 0.2. 0. and 1.1.5 and intercept of 0. 0.25. (1) Select one sample with a normal. then the limit for the test system would be 0.90. 90.” The lowest available sample should be used. Because it is the lowest point on the linearity scale. in a laboratory performing reticulocyte assays on bone marrow transplant patients.50. and Maston. 75.15%.1).” Select a second sample with a reticulocyte count that is at the lowest level of reticulocyte counts routinely assayed in the laboratory (~5 x 109/L) and label it “Sample L.50. 10. and 100. precision. and 0 mL of dilute Sample L into tubes 0 through 100.10.0% reticulocytes (see Section 9. 0. 50. because the exact definition of a reticulocyte is a function of the assay. All four of these test criteria must be tested and found acceptable before determining the sensitivity of the device. This step permits accurate pipetting of nonviscous fluid with variable-volume. then resuspend in ten volumes of PBS or any compatible buffered salts solution (~50 mL). then pipette 1. All rights reserved.e.060 mL (60 µL) An NCCLS global consensus guideline. and the previously established performance limit for the NMB method is 0. It is dependent upon trueness. The equations derived from accuracy analysis can be used to determine precision at the level of the assay equivalent to 0. It is also necessary to remove any traces of incompatible plasma. Schimenti. (2) (3) 20 .00. This becomes a system of successive approximations. remove PBS. For example. and necessary. the following procedure should be used. For example. the numerical limit to which a particular system might need to perform may be different. 0. the reticulocyte count is ~ 0. respectively. Lacerna. and 100. if the test system correlates linearly with the NMB method with a slope of 0.Number 8 NCCLS 8. 21 Centrifuge both blood samples and remove the plasma. Thus. Typically.1% or less.. i. Mix each tube thoroughly. Determination of the actual required limit of clinical performance can only be made after reference ranges have been established and after extensive experience with the test system.4 Sensitivity of Testing Sensitivity of flow cytometric reticulocyte analysis is the ability of the method to detect and reliably quantitate low percentages of reticulocytes. 0. respectively. and method bias. air-displacement pipettes.1% levels. it might be desirable. ©NCCLS. 10. 0.

Example data have been synthesized but are similar to measured data. 21 . and 10% intervals of dilution. 0. hr = the absolute reticulocyte count of Sample H (high count sample).87 1.25. One can expect randomly distributed residuals and r2 at > 0. Example of Linearity Data* Fraction of High Retic Blood (a) Retic (Ear) 0. Panel B. (4) After obtaining the results. and r2 value. All rights reserved. lr = the absolute reticulocyte count of Sample L. the sensitivity of the test system has been estimated.5 0.Volume 24 H44-A2 of ABO-compatible plasma (or fetal bovine serum) to each tube.3 0.10 0. and lt = E%r = the reticulocyte percentage.10 19. Linear regression is performed. Eat = the expected absolute total RBC count (x 109/L).30 3.75% reticulocytes).98.98.5 0.1% reticulocytes). Panel A. the absolute total RBC count of Sample L. One example of possible nonlinear data (synthesized) is shown in Figure 3. * The data from Table 4 is graphed in Figure 3.00 150.500 4. it is necessary to perform an independent absolute red cell count and convert percent reticulocyte counts to absolute reticulocyte counts.8 0. 0. Dashed lines are the best fit lines. The expected absolute reticulocyte counts are then used as the independent variable and the measured values as the dependent variable (the 0 and 100 tube counts are taken to be the truth). calculate the expected absolute values using Equations 15 and 16 and the expected percentage reticulocytes using Equation 17. low reticulocyte count of 5..1.1.90 135. If there is considerable nonrandom distribution of the residuals and if r2 is lower than ~0. For example. Mix thoroughly. NOTE: On instruments where absolute counts are not performed. ht = the absolute total RBC count of Sample H. At that point.a)lr Eat = aht + (1 . The synthesized data in B are for a high reticulocyte count of 60.g.40 0.250 4.25 41.a)lt E%r = Ear / Eat x 100 where: Ear = the expected absolute reticulocyte count (x 109/L).900 4.5 1.72 2.75 113. Linearity and sensitivity can be checked by examining the residuals. ©NCCLS.0 (15) (16) (17) Expected RBC (Eat) 5..100 4. then Table 4 shows the results of calculations using Equations 15 through 17.0 x 1012/L RBC and 150 x 109/L (3. An NCCLS global consensus guideline. slope and intercept.00 5. then Sample L may be below the level of sensitivity.50 77. then aliquot the appropriate amount from each tube to the test assay system. a = the fraction of the high reticulocyte blood in the tube (e.0 0.68 3. Table 4. The test can be repeated with more dilutions to determine where deviation from linearity occurs.750 4.. 0.000 %Retic (E%r) 0. if Sample L has 5 x 1012/L RBC and 5 x 109/L reticulocytes (0.000 4.75 Measured Retic 5 20 35 80 120 140 150 NOTE: Counts are x 109/L.00). Ear = ahr + (1 . and Sample H has 4.

with any carry-over being of no clinical significance in a properly operating instrument. after service that could affect hydraulic or pneumatic operation. particulate or cell carry-over. In those cases where the assay is internally controlled. samples used for examination of carryover should be selected for their elevated and depressed RBC counts and not for reticulocyte count values. a low sample may be created by dilution of red cells with autologous plasma. Clearly. ©NCCLS. this is represented by the RBC event (count) data. Carry-over is expressed as the percent effect of one sample upon succeeding analysis. .. a negative effect of a low-sample analysis on a succeeding higher result. this test should examine the directly measured parameter containing the largest data set.5 Testing for Carry-Over The effect of one sample upon the immediately succeeding sample result should be minimal. instruments that use fluorescent dyes must have test procedures to ensure that dye carry-over does not affect the measured intensity of the mature RBC population in a subsequent unstained sample if unstained samples are required to control the assay. To be most sensitive. after routine maintenance. In closed systems in which absolute counts are being reported. The first is carry-over of reagents. it is necessary to demonstrate that dye carry-over does not affect other assay types run subsequent to reticulocyte analysis. Linearity 8. There are two forms of carry-over that can affect data in subsequent analyses.e. If necessary. In the case of flow cytometers. i. which are used to create the characteristic signature of the particles under analysis. and then periodically validated during normal use. The status of sample carry-over should be verified at the time of initial installation. and an elevated sample can be made by centrifugation of a sample and removal of a portion of the plasma. All rights reserved. the second is a distinct form of carry-over. It is suggested that the elevated RBC count sample be selected with the value above 6 x 1012/L and the low RBC count sample be selected with the value below 2 x 1012/L.Number 8 NCCLS Figure 3. or a positive effect of an elevated sample on a succeeding lower result. 22 An NCCLS global consensus guideline. Carry-over can exert a negative dilution effect of diluent from a rinse cycle transported into the first sample analysis. Systems using nonfluorescent technologies must demonstrate that reagent carry-over does not affect assays of other types run on the same instrument. Reticulocyte instrument analysis is tested for carry-over by testing the effect of an elevated sample on a succeeding low-sample result.and instrument-specific. The reagent carry-over is methodology.

28.0. The advantages of the commercial control An NCCLS global consensus guideline.5 However. (Please refer to the most current version of NCCLS document C24— Statistical Quality Control for Quantitative Measurements: Principles and Definitions for additional information. Aliquots are stored at 2 to 6 °C and are counted at regular intervals for up to three weeks.1 9. but it must not exceed a period of 72 hours.73% (6. and i3 are consecutive analyses of the elevated RBC sample and j1. and they should be warmed to room temperature immediately before each analysis. Whole blood (human or rabbit) collected in citrate phosphate dextrose (CPD) has been proposed as a quality control. if required) throughout the day or 72-hour period to determine that there was no significant shift (>3 SD) in instrument performance.Volume 24 H44-A2 Analyze an elevated RBC count sample three times consecutively.90. the QC procedure should include a minimum of two levels of control material. QC analyses should be run at an interval within which the trueness and precision of a testing system is expected to be stable.) This document is limited to the QC of the analytic process. Samples should be reanalyzed (and restained. 0. 6. Determine carry-over percent by use of the following statistical formula: Carry-over (%) = (j1 .25. Quality control is divided into long-term QC and short-term QC. All samples used for this purpose should be stored at 2 to 6 ºC between runs. Several commercial reticulocyte control preparations are also available from various vendors and use should follow the manufacturer’s instruction. i2.1. and j3 are consecutive analyses of the low RBC sample. The reticulocyte count decreases during storage.90 .j3 ) (i 3 .87.86 (j) samples. as no such data exists in the published literature. It is recommended that QC material(s) be run once within each eight hours of operation. j2. In addition.0.86) Carry-over = Compare the results of this carry-over testing procedure with the stated instrument performance specifications.31 . followed by three consecutive analyses of the low RBC count sample. Example: Consecutive results from analysis of high 6. apparently at a predictable rate. repeated warming cycles of such material should be investigated by either the manufacturer or laboratory. preferably in both the abnormal and normal ranges of reportable patient values. 0. ©NCCLS. This involves collection and treatment of the specimens. The manufacturer's recommendations regarding the inclusion of controls and calibrators (if available) should be followed. this control system has not yet been widely evaluated for effectiveness.86) x 100 = 0.j3 ) x 100 (18) where i1.1 Short-Term Quality Control Procedures Red Cell Samples One or more patient samples may be selected from the first analytical run after assurance of acceptable instrument performance by use of the long-term control procedure. the analytical process. 23 . 22 9. However. and reporting the results. A reticulocyte count is determined shortly after collection. 6. All rights reserved. (0.31 (i) and low 0. 9 Quality Control and Quality Assurance Quality control (QC) is an organized system for continuously monitoring the analytical processes to ensure true and precise data.

Use of only one method is necessary to detect instrument performance drift and ensure consistent instrument operation over an extended time period. independently assigned assay values. ©NCCLS. reticulocyte count are prepared according to the method outlined in Section 6. including reticulocyte counting. Differential Leukocyte Counting. Additionally.Number 8 NCCLS material include a longer stability than whole blood material. It is important for laboratories to verify the 24 An NCCLS global consensus guideline. 9. and each time a new reagent is introduced.000 consecutive red cells are examined for blue-staining granules. the long-term control material should be analyzed after instrument service that affects hydraulic.2.1. IL: College of American Pathologists. . Table 5. i. an elevated. Using this method. preferably. The mean percent of reticulocytes of both blood films is determined. 1978:39-45. All rights reserved.. the process is dependent upon availability of suitable control material. In: Koepke JA. are supplied with predetermined assay target and limit values.2 NMB-Stained Blood Films For QC purposes. Commercially available materials for use in hematology analyses.) Instrument Count (%) Tolerance Limits (%) 0 1 2 3 4 5 6 7 8 9 10 0-1 0-2 1-4 2-5 2-6 3-7 4-8 5-9 6-10 7-11 8-13 9. or electronic operation. after periodic maintenance. Long-term QC is usually done by multiple analyses over many days of an aliquot of stabilized or stored blood. reticulocytes.2 Long-Term Quality Control Procedures Although the long-term QC procedures are intended to monitor and ensure the stability of instrument performance. pneumatic. 1. 95% Confidence Limits of NMB Reticulocyte Count (Adapted with permission from Rümke CL.1 Stabilized Control Materials Control material that is stable for one month or longer should be measured at least once at the beginning of each work run in which the instrument accuracy and precision is demonstrated to be stable. These limits are comparable to intervals calculated by the binomial method. two blood films from a sample with a normal or. most often from a commercial source. and the potential to be included in interlaboratory QC programs. The long-term process controls described below employ either commercially available control materials or controls produced in the laboratory. The statistically expected variability in differential leukocyte counting.e. 9. Skokie. ed. The reticulocyte counting system is performing acceptably if the blood film reticulocyte counts are within the limits listed in Table 5.

10. the tolerance limits are recalculated around the mean of 20 or more values. and equally divided between men and women. at least 40 normal adult samples are required. These tentative control limits. Studies by several subcommittee members have noted small but significant differences between the reference ranges of men An NCCLS global consensus guideline. Differences in reference ranges between the reference method (reference measurement procedure) and flow cytometric methods can be anticipated if there is a bias in the slope and/or in the linear regression intercept (see Section 8.1. Control material manufacturers usually produce each lot of material with values near the same target value.Volume 24 H44-A2 assigned target values before use when assays are provided for their instrument systems. all QC protocols should encompass any additional parameter(s) that are reported by the test instrument. including red and white cell parameters. including reticulocytes (percent). and any parameter unique to the test device.) These persons should have a normal blood count. ©NCCLS. and ±3 SD ranges around the mean. The control target value is then determined by calculation of the arithmetic mean determined from the ten or more analyses. or acceptable range (±2 SD).1 Reference Population At least 100 normal adults between the ages of 18 and 50 years. ±2. A minimum of ten data points is required. Another method for development of control limits is the use of historical data to establish a set of permanent limits that will be used with each new lot of control material. Consistency of these target values allows the laboratory to use historical ±2 SD limits to establish a set of stable limits that can be used continuously. If a reference range has not been established for the comparative method. In the second phase. Establishment of the performance limits for the control material is accomplished in three phases. apparently due to incorrect use of the Miller disc. Control material target value determination should be done with a properly calibrated and maintained instrument. Each laboratory using flow cytometric reticulocyte counts should establish reference ranges for its particular method. RBC count. 23 In addition. 25 . The target value should not be determined on less than ten data points or from analyses done in less than five days. 10 Reticulocyte Reference Intervals and New Reticulocyte Parameters Reference intervals for each system (automated reticulocyte counting instrument/dye) can be different and. should be established for each instrument/dye combination. The first phase establishes the tentative target value from the initial ten analyses. calculate ±1. Analyze each level of the material twice a day (preferably at different times during the workday) for a minimum of five days to establish the target value. concurrent with the determination of the target value (mean of ≥10 values).” Biases have been noted. therefore. The same samples may be used for determination of both reference intervals. (See the most current version of NCCLS document C28—How to Define and Determine Reference Intervals in the Clinical Laboratory. It is important to know the method(s) used by the manufacturer for assigning reticulocyte counts to “controls with assigned values.24 For confirmation of a system reference range. it is appropriate for the laboratory to do so in conjunction with the instrument reference interval study. are required for the determination of reference ranges. This new set of control limits is then used for the remainder of the life of the particular lot of control material. and the data may additionally be used for the accuracy validation study (see Section 8. unless the target value changes significantly or the instrument performance is altered.4). All rights reserved.1). are then used until sufficient additional data are collected. In the third phase.17. immature reticulocyte fraction. or to establish values that are specific for their laboratories and instrument systems when predetermined assay targets are not provided.

10. visual methods to quantitatively describe the reticulocyte maturational stage in the peripheral blood are usually variations of the work of Heilmeyer28 and are not sufficiently reproducible or precise to routinely offer in clinical laboratory practice.3 Immature Reticulocyte Fraction (Reticulocyte Maturation Index) Automated reticulocyte counting has provided laboratories with improved precision in reticulocyte enumeration. It has also increased an interest in the clinical utility of reporting an additional reticulocyte parameter quantitating the immaturity level of a reticulocyte population.2 Reticulocyte Reference Intervals Due to methodologic differences and the lack of absolute reticulocyte standards. Reference ranges should be expressed in terms of absolute reticulocyte count. each laboratory should determine. 25 Table 6. not a mean +2 SD.29 Flow cytometry reticulocyte counting instruments provide the ability to analyze with acceptable precision the distribution of staining intensity (or similar physical or chemical measures) of the RNA-binding or visualization reagents. 26 manual/visual Ethidium bromide Auramine O26 Thiazole orange Thioflavin T Adults Newborns Oxazine 750 Cell Dyn CK52026 48 50 50 70 65 130 75 55 20-110 20-85 27-95 15-135 30-100 65-230 20-135 26-84 9 *This table also contains unpublished data from several committee members. or at least confirm. which is directly proportional to the amount of cellular RNA. Therefore. Results are expressed as the number of reticulocytes x 109/L. should be taken as the reference range. Reference ranges for reticulocytes stained with fluorochromes tend to be somewhat higher than those done with vital dyes. but establishment of laboratory specific ranges in the various pediatric age groups may not be practical or feasible due to ethical constraints. reference ranges are dependent upon the dye or fluorochrome being used (see Table 6). will differ from adult reference ranges. . studies have shown a significant biological diurnal variability of the reticulocyte count. Therefore. Fluorochrome/Dye Mean 95% Range * New methylene blue. except as referenced. 10. All data is given as number of reticulocytes x 10 /L and is rounded to the nearest 5. The data are almost invariably log-normally distributed. 26 An NCCLS global consensus guideline.Number 8 NCCLS and women. All rights reserved. quantitative assessment of the distribution of the highly fluorescent or highly stained reticulocytes can provide a sensitive indicator of erythropoietic bone marrow activity.27 Attempts by the laboratorians to describe and quantitate reticulocyte maturation can be traced back to the work of Ehrlich. Because the population of reticulocytes most recently released from the bone marrow contains the highest concentration of RNA. particularly in the ages of <2 years. Reporting reticulocyte counts solely as a percentage is archaic and not acceptable clinical practice. Tentative Reticulocyte Reference Intervals2. the central 95% of results. ©NCCLS. its own reference ranges. Results are from adult males and females with no statistically significant differences between sexes observed. This variability is approximately ± 20% of the mean reticulocyte count over a period of four weeks. Pediatric reference ranges.6 However. Also.

6.Volume 24 H44-A2 Since the last edition of this document.14. individual laboratories should establish or verify manufacturer’s suggested reference ranges based on their own performance characteristics. this will drive the need for means of quality control and calibration of measurements of hemoglobin levels in reticulocytes.S. All rights reserved. “immature reticulocyte fraction (IRF).25.50. comparable to other methods. 27 . only one manufacturer offers measurement of CHr but other related parameters have also been reported to correlate with iron deficiency or hypochromic red cells. Clinical Utility of Immature Reticulocyte Fraction Monitor Bone Marrow or Stem Cell Regeneration Timing for Stem Cell Harvests Following Growth post BMT or ChemoRx Factor or Cytotoxic Drug Therapy Monitor Neonatal Transfusion Needs and Verify Aplastic Anemia Prognosis in Anemia of AIDS and Prematurity Monitor Renal Transplant Engraftment (Epo Monitor Bone Marrow Toxic Insults from Drugs Production) (e.”31 “RNA content. the U. such as the RBC-Y and Ret-Y measurements on another automated reticulocyte analysis instrument. because instrument and reagent performances and reference ranges differ. 52 If clinical utility of these parameters is confirmed with continued experience. Myelodysplasia. consensus on the terminology measuring reticulocyte maturity has been reached through effort of the International Society for Laboratory Hematology (ISLH). However. Davis and Bigelow introduced the term “reticulocyte maturity index (RMI).41-46 Some automated analyzers are capable of measuring the mean corpuscular volume of reticulocytes (MCVr). 39. clinicians. in early studies using thiazole orange (TO) flow cytometry methodology. AIDS.15.” expressed in arbitrary mean fluorescence intensity units. Infants. Blood Donation Monitor Efficacy of Anemia Therapy Evaluate Normochromic Anemias of Various Etiologies Detection of Aplastic Crisis in Hemolytic Anemias Detection of Occult or Compensated Hemorrhage or Hemolysis 10. The IRF is a promising new hematology parameter that needs further integration into clinical practice with cooperation between manufacturers. AZT) Classification of Anemias Monitor Erythropoietin Therapy: Renal Failure. 35-38 Other instruments have been developed that measure the percentage and number of reticulocytes that fall into arbitrarily defined regions of the data presentation histogram from which an IRF. An NCCLS global consensus guideline. ©NCCLS.51 Recent publications indicating measurement of hypochromatic reticulocytes has clinical utility in identifying states of functional iron deficiency may drive a clinical demand for expanded automated measurements of reticulocyte parameters equivalent to the CHr or Ret-Y. and standards-setting organizations.6.”32 “highly fluorescent reticulocytes. In practice it has effectively replaced the previous practice of adjusting the reticulocyte count based upon the level of anemia.4 Reticulocyte Hemoglobin Content (CHr) The reticulocyte hemoglobin content (CHr) is reported to approximate the frequency of hypochromic reticulocytes in blood. both manufacturers and clinical practice have accepted the term proposed by the ISLH consensus process. These terms include “RNA index. 30 Various other terms have been used in the literature to describe reticulocyte immaturity measurements. laboratorians.”33 and “immature reticulocyte fraction.” 34 Since the last edition.35.40 Since the last edition of this document. The utility of the IRF has been validated in the literature for a variety of clinical applications. can be derived. Food and Drug Administration (FDA) has cleared the IRF for diagnostic use on at least three manufacturer’s instruments. and clinical use has progressively increased.g.. as summarized in the table below.47-49 At present.38 Table 7.

Iaffaldano C. Sutherland. ISO 15193. 1996. et al.102:468-477. 1992. Nagao A. 1994. 2003. Cytometry. Reticulocyte analysis: comparison of flow cytometry. Davis BH. 1989:16. College of American Pathologists’ Reticulocyte proficiency testing program using surrogate blood: Insights into contemporary clinical practice. Amer J Clin Pathol. London: Arnold. International Vocabulary of Basic and General Terms in Metrology. Tsuda I. ISO. DE. Geneva: International Organization for Standardization. Houwen B. MN. A time-saving device for the counting of reticulocytes. Jones AR.13:407-415. Twedt D. Application of the R-1000 automated reticulocyte analyzer to research on in vitro reticulocyte maturation. Gilmer PR.106:723727. Reticulocyte counting: methods and clinical applications. Subsection 1217 (b) (2):7166.13:853-862. 1998. Spier CM. Davis BH.Measurement of Quantities of Quantities in Samples of Biological Origin . In Vitro Diagnostic Medical Devices – Measurement of Quantities in Biological Samples – Metrological Traceability of Values Assigned to Calibrators and Control Materials. Minneapolis. Sysmex J. 4th ed. Reticulocytes in human preserved blood as control material for automated reticulocyte counters.3:83-90. Expert Panel on Cytometry. et al. All rights reserved. Statistics-Vocabulary and Symbols – Part 1: Probability and General Statistical Terms.16:157-174. Am J Clin Pathol. . Schneiderman M. International Council for Standardization in Haematology (ICSH). Rules and Regulations. Krent A. eds. Am J Clin Pathol. ed. 2002. Kinetics of erythrogenesis after bone marrow transplantation. Reticulocytes. Lab Hematol. Preston FE. van Assendelft OW.20:77-79. Practical Flow Cytometry. 1997. Swaim WR. Clin Lab Haem. Koepke JA.Presentation of Reference Measurement Procedures. 1992. Flow cytometric reticulocyte analysis: Multi-institutional interlaboratory correlation study.102:468-477. Moeschberger ML. Subpart K. International Committee For Standardization in Haematology. In Vitro Diagnostic Systems . 1990. Omaha. Package inserts: Retic-Chex7 Streck Laboratories. 1992. 2002. 1993. Fujimoto K. Part 493. Statistics-Vocabulary and Symbols – Part 1: Probability and General Statistical Terms. Reticulocyte maturation. Bigelow NC. Proposed reference method for reticulocyte counting based on the determination of the reticulocyte to red cell ratio. Rowan RM. 1992.18:167-186.20:1079-1083. Koepke JA. Tatsumi N. ISO 3534-1. Lazarus HM. Bigelow NC.8:169-179. Guidelines for the evaluation of blood cell analysers including those used for differential leucocyte and reticulocyte counting and cell marker applications. Federal Register. Clin Lab Haematol. The reticulocyte: An approach to definition. Kachin J. 1986. Advanced Laboratory Methods in Haematology. d’Onofrio G. Geneva: International Organization for Standardization. et al. Maston L. Guidelines for reticulocyte counting by microscopy on supravitally stained preparation. Amer J Clin Pathol. 1993. NY: John Wiley & Sons.38:648-650. Blood Cells. Edinburgh: Churchill Livingstone. Amer J Clin Pathol. ISO. image analysis.94:109-110. NE. Chanarin I. ©NCCLS. Schimenti KJ. Clin Lab Haematol. Geneva: International Organization for Standardization. 2002:78-126. Am J Clin Pathol. 1994. International Committee for Standardization in Haematology. Retic-I7 R&D Systems. Shapiro HM. Geneva: International Organization for Standardization. New York. WHO LBS/92. Gottfried E. Cornbleet PJ. and manual counting. Laboratory Hematology. Estimation of reference ranges: How many subjects are needed? Clin Chem. 1993. Am J Clin Pathol.97:574583. Diurnal change of blood count analysis in normal subjects.66:262-267.Number 8 NCCLS References 1 Koepke JF. Zini G. 1986. Koepke JA. In: Rowan RM. Nakamura S. 1990. Lott JA. Davis BH. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 28 An NCCLS global consensus guideline. Lacerna K. Koepke JA. 1992. Flow cytometric reticulocyte analysis: Multi-institutional interlaboratory correlation study. Geneva: International Organization for Standardization. Brecher G. et al. ISO 3534-1. Geneva: World Health Organization.3. ISO/FDIS 17511. 1950. 1994. ISO. Mitchell LC. ISO.

et al. Limpanasithikul W.Volume 24 27 H44-A2 Davis BH. Sorette M.93:7078. Huhn D. Arch Pathol Lab Med. Davis BH. 1994. 1996. Arch Pathol Lab Med. 1993. a useful clinical parameter of erythropoietic activity. Kuse R.96:823-833.109:325-329.108:138-142.38:148-159. 1989. Bulian P. All rights reserved. Cremins J. 1997. Cytometry. Bone Marrow Transplant. Roger R. and manual counts. flow cytometry. Effects of subcutaneous recombinant human erythropoietin in normal subjects: Development of decreased reticulocyte hemoglobin content and iron-deficient erythropoiesis. et al. Greinix HT. Fishbane S. Am J Clin Pathol. 1990. Immature reticulocyte fraction (IRF) and reticulocyte counts: Comparison of Cell-Dyn 4000.21:829-840.115:100-111. shift. Driessen A. Evaluation of the Sysmex R-1000: An automated reticulocyte analyzer. Davis BH. New red cell parameters on the Sysmex XE-2100 as potential markers of functional iron deficiency. 2001. Gratwohl A. and shift reticulocytes. Venudo A. Lab Hematol. et al. Rizzotti P. Automated reticulocyte counting and immature reticulocyte fraction measurements: Comparison of ABX Pentra 120 Retic. et al. et al. Inter J Sports Med. Van Hove L. DiCorato M. Diagnostic utility of red cell flow cytometric analysis. Flow cytometry of reticulocytes applied to clinical hematology. Seminars in Hematology. Charuruks N. Colella GM. Reticulocyte hemoglobin content in the evaluation of iron status of hemodialysis patients. Tichelli A. Immature reticulocyte fraction (IRF): By any name. thiazole flow cytometry. Hahn AG. Galgano C. iron.2:144-150. Kaplow L. Bigelow N. Parisotto R. Sysmex R-3000. et al. et al.123:660-667. J Lab Clin Med. Laboratory evaluation of the Miles H3 automated reticulocyte counter: A comparative study with manual reference method and Sysmex R-1000.21:471-479. Proposal for standardization of flow cytometric reticulocyte maturity index measurements. Westhaeuser R. 1995. Brugnara C. Bulian P. Blood. et al. Zelmanovic D. Sysmex J Int. Lacombe F. Skikne B. Reference ranges of reticulocytes in adults.102:471-479. et al. Hipp MJ. Brugnara C. Goodnough LT. Vial J-P. Buttarello M. Am J Clin Pathol. Blood. 1985. Ztschr Klin Med. 1993. Clin Lab Haematol. Am J Clin Pathol. Farina G.37:93-130. Bigelow NC. Davis BH. 2000. 2001. 1999. Infus Ther Transfus Med. Brugnara C. J Med Assoc Thai. Briggs C. Erythropoietin. Langley RC Jr. Thompson B.61:1091-1097. Arch Pathol Lab Med.28:256-262. ©NCCLS. The appearance of reticulocytes with medium or high RNA content is a sensitive indicator of beginning granulocyte recovery after aplasiogenic cytostatic drug therapy in patients with AML. Sysmex R-2000. Machin SJ.66:213-214. 1993. Diagnostic advances in defining erythropoietic abnormalities and red cell diseases.112:677-686. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 An NCCLS global consensus guideline. Tanke H. Flow cytometric reticulocyte counting – Parallel evaluation of five automater analyzers: An NCCLS-ICSH approach. Rothbarth P.14:318-326. 29 . Flow cytometric reticulocyte quantification using thiazole orange provides clinically useful reticulocyte maturity index. 1944. 1932. Serke S. Linkesch W. 1983. Crit Rev Clin Lab Sci. Reifungsstadien an Überlebenden Reticulozyten In Vitro und ihre Bedeutung für die Schaetzung der täglichen Haemoglobin-Produktion In Vivo. The analysis of iron status in clinical laboratories.14:307-313. Gore CJ. cells. Brugnara C. Bigelow NC. 1997. 2001. Keil F. 2001:11(2). 1998. Buttarello M. Heilmeyer L. Davis BH. Reticulocyte hemoglobin: An integrated parameter for evaluation of erythropoietic activity. 2000. Reticulocyte parameters as potential discriminators of recombinant human erythropoietin abuse in elite athletes.2:2-8. Relationship of reticulocyte age to polychromasia.113:684-689. 1996. and manual counts. Lab Hem. Early detection of hematopoietic engraftment after bone marrow and peripheral blood stem cell transplantation by highly fluorescent reticulocyte counts. Kidney Int. Voravud N.81:356-364. Ann Hematol. Langweiler M. Irving PJ. Schaefer M. 2000. Reticulocyte cellular indices: A new approach in the diagnosis of anemias and monitoring of erythropoietic function. Davis BH. Vossen J. Crouch J. Am J Clin Pathol. Improved specificity of determination of immature erythrocytes (reticulocytes) by multiparameter flow cytometry and thiazole orange using combined staining with monoclonal antibody (anti-glycophorin-A).15:33-44. Clinics in Laboratory Medicine. Lacoste L. and erythropoiesis.121:361-365.52:217-222. 1994. 2001.119:1141-1148. Reinecke T. Automated reticulocyte counting and measurement of reticulocyte indices. et al. Brugnara C. Am J Clin Path. Evaluation of the Miles H*3 blood analyzer.

Section 3. For further information.nccls. A definition has been added to the document. CD71 is the current accepted term for what was previously referred to as “transferrin receptor. “Hypochromic red cells” is too specific a descriptor for a key word search. • 4. A reference should be included for the term “static cytometry. the suggested additions have not been incorporated. • 6. Also. the use of Thiazole Orange should be extended. . The new title does not limit the document to flow cytometers and.” Therefore. contact the Executive Offices or visit our website at www. Approved Guideline—Second Edition Key Words 1. Definitions (Formerly Section 5) 3. this terminology would be included in key word searches for “red cells” and/or “anemia. • 7. ©NCCLS.” Why is that? Auramine O is still a valid example and should be included. b) RBC-Y is the mean value of the forward light scatter histogram of mature red blood cells. Both Thiazole Orange and Auramine O are listed as examples. All rights reserved. Principle (Formerly Section 4) 2. • 30 An NCCLS global consensus guideline. Summary of Delegate Comments and Area Committee Responses H44-A2: Methods for Reticulocyte Counting (Automated Blood Cell Counters. Is Anti-CD71 a better example than Transferrin? We suggest including the original and new examples. The definition that appears in this version of the guideline is current and accepted. Flow Cytometry. d) contains an incomplete sentence.” • Section 4. In this paragraph the examples have changed from the previous publication: “Auramine O” to “Thiazole Orange” and “Transferrin” to “Anti-CD71. The definition of “accuracy” has been revised. and Supravital Dyes). Please add the following definitions: a) % Hypochromic red cell (% Hypo) is a percentage of red blood cells with lower than normal hemoglobin content. therefore. • Please add the following key words: RBC-Y. The definition has been added.” Which term follows the NCCLS definition? That is the one to use. A definition of “reference method” should be included. c) RET-Y is the mean value of the forward light scatter histogram of mature reticulocytes. A definition of the term “immature reticulocyte fraction” should be added to this section. Accuracy b) “Accepted reference values” is a new term that is introduced and needs a definition. See the response to Comment 1. Two of the terms are specific to one manufacturer’s instrument and are not accepted medical terminology.org. • 5. These terms are not accepted key words for literature searches.Number 8 NCCLS NCCLS consensus procedures include an appeals process that is described in detail in Section 8 of the Administrative Procedures. • 8. and % hypochromic red cells. The previous version of this document (H44-A) used the term “replicate measurements” instead of the newly introduced phrase “independent test results obtained under prescribed/stipulated conditions. RET-Y.” A reference has been included.

to support the change. ICSH calls this a proposed reference method. not experimental evidence. Please provide a reference for the change. • The change from a minimum of 20 blood samples to a minimum of 26 blood samples is based upon a statistical recommendation. Please correct if necessary and/or explain the changes.1.1. 16. Now it is changed to 26. 1998. The text has been maintained in Section 6.3.1.1.4 provides details specific to reticulocyte analysis. Specimen Handling and Packaging (Formerly Section 6. as the documented imprecision and subjectivity of the method yield it unacceptable. 31 . • H44-A2 Obtaining the “reason why the test is requisitioned” will be difficult and impossible to obtain. The last sentence belongs in Section 6.1.20:77-79. Table 2. Sample Evaluation (Formerly Section 6. the numbers in A2 appear to have a two decimal place difference from the table in A1. • A reference to the H18 guideline has been added as recommended. All rights reserved.2) 10. ©NCCLS. The terms “correlation” and “comparability” have been substituted for “reference. particularly in light of the existence of methods demonstrating superior performance.2.1.5) 11. However. Section 5.1) 14. References are not available.1. the Retic % was . • This section is intended for instrument evaluations and periodic validation of instrument performance if calibration problems are suspected. Section 8. Under the SE p column. • The rationale for the suggested modification is not clear.5. The 95% SE p has been deleted in the table in A2. Section 6. Shouldn’t the NCCLS document state this as proposed reference method and not use the terms correlation or comparable? • The current ICSH committee does not favor NMB as a reference method (reference measurement procedure).2.4. Why is this? According to Clin Lab Haem. For example. This section is not intended to provide recommendations for routine or daily quality control functions.” In some countries.2. It was recommended previously that a minimum of 20 blood samples be examined.1) 9. Is this section necessary since there are other NCCLS documents that address this issue? • The subcommittee believes this section is a necessary component of the guideline.. this information is required to justify billing claims. Section 5. Please include reference to NCCLS document H18—Procedures for the Handling and Processing of Blood Specimens for completeness. Correlation Method-New Methylene Blue Visual Microscopy (Formerly Section 7) and Section 8.Volume 24 Section 5.4) 12. The 95% Low and High in the original table have been substituted for Upper and Lower Limit in the table A2.3. The method to evaluate reticulocyte counting by flow cytometry is complete and well described. Patient Information (Formerly Section 6. nor needed.” which was in the previous publication. We suggest that this statement be deleted. Section 5.1.2.1) 13. and Section 5. so deletion of the statement would not be appropriate. Section 8. “…ideally should state the reason why the test is requisitioned. Other NCCLS documents contain general information on Sample Evaluation.3) 15. it is difficult to perform all of the steps for confirmation of accuracy and precision of automated analyzers as they are too time consuming. An NCCLS global consensus guideline. Comparability (Correlation) Method (Formerly Section 9. Accuracy Estimation: Paired Sample Method: It appears that there are several differences from the previous table in H44-A. such as the U. (Diagnostic Sensitivity Studies (Formerly Section 7.01 in the original table and now it is 0.S. Specimen Collection Techniques (Formerly Section 6. The text has been modified to read. Procedure (Formerly Section 7.

19. Nephrol Dial Transplant. 1997:12:1173-1181.4) 18. iron supplementation in maintenance haemodialysis patients. Lindner K. Section 8. 32 An NCCLS global consensus guideline. This sample size will allow for a confidence interval of 95%. . Some additional methods related to reticulocyte analysis have been incorporated in Section 10.5) 17. Heidler RA. Reinecke T. Rogers R. The RBC-Y/RET-Y has been reported to be used as sensitive and specific indicators of functional iron deficiency. Machin SJ. % Hypo has been reported to be used as sensitive and specific indicators of functional iron deficiency in clinical studies. Horl WG. All rights reserved. 46 • The references have been added as suggested. 2001. Reticulocyte Hemoglbin Content (CHr) (Formerly Section 11.4. Please add the following reference: 47 Braun. Infus Ther Transfus Med. New Red Cell Parameters on the Sysmex XE-2100 as Potential Markers of Functional Iron Deficiency. ©NCCLS. 2 (2001). No. 28:256-262. Testing for Carry-Over (Formerly Section 7. Please add the following references: 45 Briggs C. • This section is based upon intuitive mathematics with definitions and explanations provided. Section 10. The Analysis of Iron Status in Clinical Laboratories. Thompson B. Is it an ICSH procedure? We recommend that other comparable methods be added after Section 11. Other methods for functional iron deficiency determination are not based upon reticulocyte analysis and are therefore. these two values seem to equate with red cell/reticulocyte hemoglobin content.v. Other changes were made to correct inaccuracies in the previous version of the document. Schaefer M. Sysmex Journal Int Vol.3.Number 8 • NCCLS Changes in this version of the guideline are based upon statistical consultation. Schreiber M. indicating that 26 samples is the preferred “N” to verify a normal range.4 has been modified and the suggested reference has been added. • The text in Section 10. outside of the scope of this document.5. According to the literature. 11. Percentage hypochromic red cells as a predictor of erythropoietic and iron response after i.4 since there are other alternatives to monitoring functional iron deficiency than just CHr. This needs a reference.

Volume 24 H44-A2 NOTES An NCCLS global consensus guideline. 33 . All rights reserved. ©NCCLS.

GP26-A2 defines a clinical laboratory path of workflow which consists of three sequential processes: preanalytical.” basic to any organization. The quality system essentials (QSEs) are: Documents & Records Organization Personnel Equipment Purchasing & Inventory Process Control Information Management Occurrence Management Assessment Process Improvement Service & Satisfaction Facilities & Safety Path of Workflow A path of workflow is the description of the necessary steps to deliver the particular product or service that the organization or entity provides. 34 An NCCLS global consensus guideline. For example. The quality system approach applies a core set of “quality system essentials (QSEs). All rights reserved. defines a document structure via a template.” For a description of the other NCCLS documents listed in the grid. The QSEs provide the framework for delivery of any type of product or service. H44-A2 addresses the clinical laboratory path of workflow steps indicated by an “X. serving as a manager’s guide. ©NCCLS. and provides a process to identify needed documents through a gap analysis. which facilitates project management. to all operations in any healthcare service’s path of workflow. All clinical laboratories follow these processes to deliver the laboratory’s services. and postanalytical. analytical. Results Report X . namely quality laboratory information. The approach is based on the model presented in the most current edition of NCCLS document HS1—A Quality System Model for Health Care. Preanalytic Test Request Patient Assessment Specimen Collection Specimen Transport Specimen Receipt Analytic Laboratory Interpretation Postanalytic Post-test Specimen Management X GP5 Testing Review X H1 H3 H4 X H18 X H18 X H20 X C24 C28 EP6 Adapted from NCCLS document HS1— A Quality System Model for Health Care.Number 8 NCCLS The Quality System Approach NCCLS subscribes to a quality system approach in the development of standards and guidelines. please refer to the Related NCCLS Publications section on the following page.

Approved Standard (1992). Approved Guideline— Second Edition (2000). to which an automated or manual test method can be compared. blood culture collection. as well as hazards to patients from inappropriate specimen collection by skin puncture procedures. plans for quality control procedures. Approved Standard—Fourth Edition (1999). and for determining and stating a manufacturer’s claim for linear range. C28-A2 EP6-A GP5-A2 H1-A5 H3-A5 H4-A4 H18-A2 H20-A * Proposed. All rights reserved. for checking linearity as part of routine quality assurance. An NCCLS global consensus guideline. Procedures for the Handling and Processing of Blood Specimens. radioactive. 35 . Approved Guideline—Second Edition (1999). regulations. Tubes and Additives for Venous Blood Specimen Collection.Volume 24 H44-A2 Related NCCLS Publications* C24-A2 Statistical Quality Control for Quantitative Measurements: Principles and Definitions. Procedures and Devices for the Collection of Diagnostic Blood Specimens by Skin Puncture. Evaluation of the Linearity of Quantitative Measurement Procedures: A Statistical Approach. readers should refer to the most recent editions. Procedures for the Collection of Diagnostic Blood Specimens by Venipuncture. Approved Standard—Fifth Edition (2003). Reference Leukocyte Differential Count (Proportional) and Evaluation of Instrumental Methods. Based on U. and venipuncture in children. This document provides guidance for characterizing the linearity of a method during a method evaluation. Approved Guideline—Second Edition (2002).S. How to Define and Determine Reference Intervals in the Clinical Laboratory. This document provides guidance for determining reference values and reference intervals for quantitative clinical laboratory tests. and heparin compounds used in blood collection devices. this document provides guidance on safe handling and disposal of chemical. ©NCCLS. sodium citrate. therefore. A consolidation of H4-A3 and H14-A2. infectious. and guidance for quality control applications. This guideline provides definitions of analytical intervals. This standard contains requirements for venous blood collection tubes and additives including heparin. Approved Standard— Fifth Edition (2003).and tentative-level documents are being advanced through the NCCLS consensus process. EDTA. Clinical Laboratory Waste Management. This guideline addresses multiple factors associated with handling and processing specimens. this standard provides detailed descriptions and explanations of proper collection techniques. as well as factors that can introduce imprecision or systematic bias into results. This document provides procedures for the collection of diagnostic specimens by venipuncture. Approved Guideline (2003). This standard describes automated differential counters and establishes a reference method (reference measurement procedure) based on the visual (or manual) differential count for leukocyte differential counting. including line draws. Approved Guideline—Second Edition (1999). and multihazardous wastes generated in the clinical laboratory.

. ©NCCLS. All rights reserved.Number 8 NOTES NCCLS 36 An NCCLS global consensus guideline.

Volume 24 NOTES H44-A2 An NCCLS global consensus guideline. 37 . ©NCCLS. All rights reserved.

Number 8 NOTES NCCLS 38 An NCCLS global consensus guideline. All rights reserved. ©NCCLS. .

All rights reserved. ©NCCLS. 39 .Volume 24 NOTES H44-A2 An NCCLS global consensus guideline.

IN Bayer Corporation . Paratek Pharmaceuticals Pfizer Inc Pfizer Inc .C. Everett Chalmers Hospital (New Brunswick. Labtest Diagnostica S. Inc.Elkhart.Marburg. Sysmex Corporation (Japan) Sysmex Corporation (Long Grove. Nissui Pharmaceutical Co. . LifeScan.San Jose. (Rochester. BC.L. Wyeth Research Xyletech Systems. Bristol-Myers Squibb Company Canadian External Quality Assessment Laboratory Chiron Corporation ChromaVision Medical Systems. Korea) Trade Associations AdvaMed Japan Association Clinical Reagents Ind. . Pfizer Inc Roche Diagnostics. CA Dade Behring Inc. AE) 67th CSH Wuerzburg.France Blaine Healthcare Associates. Inc. duPont Hospital for Children (DE) Allina Health System (MN) Anne Arundel Medical Center (MD) Antwerp University Hospital (Belgium) Arkansas Department of Health Armed Forces Research Institute of Medical Science (APO. LTD (LA) Akershus Central Hospital and AFA (Norway) Albemarle Hospital (NC) Alfred I.Methodist Hospital (IN) CLSI Laboratories (PA) Community Hospital of Lancaster (PA) Community Hospital of the Monterey Peninsula (CA) CompuNet Clinical Laboratories (OH) Cook Children’s Medical Center (TX) Cook County Hospital (IL) Covance Central Laboratory Services (IN) Creighton University Medical Center (NE) Danish Veterinary Laboratory (Denmark) Danville Regional Medical Center (VA) Department of Health & Community Services (New Brunswick. .M. BD BD Consumer Products BD Diagnostic Systems BD VACUTAINER Systems Beckman Coulter.Cupertino. Inc. Clinici ItalianiA. F. Joseph Warquignies (Belgium) Christus St. Laboratory Specialists. Ctr. Inc.Sacramento. . CA David G. Ontario Medical Association Quality Management Program-Laboratory Service RCPA Quality Assurance Programs PTY Limited Sociedad Espanola de Bioquimica Clinica y Patologia Molecular Sociedade Brasileira de Analises Clinicas Taiwanese Committee for Clinical Laboratory Standards (TCCLS) Turkish Society of Microbiology Government Members Armed Forces Institute of Pathology Association of Public Health Laboratories BC Centre for Disease Control Centers for Disease Control and Prevention Centers for Medicare & Medicaid Services Centers for Medicare & Medicaid Services/CLIA Program Chinese Committee for Clinical Laboratory Standards Commonwealth of Pennsylvania Bureau of Laboratories Department of Veterans Affairs Deutsches Institut für Normung (DIN) FDA Center for Devices and Radiological Health FDA Center for Veterinary Medicine FDA Division of Anti-Infective Drug Products Iowa State Hygienic Laboratory Massachusetts Department of Public Health Laboratories National Center of Infectious and Parasitic Diseases (Bulgaria) National Health Laboratory Service (South Africa) National Institute of Standards and Technology New York State Department of Health Ontario Ministry of Health Pennsylvania Dept. Norfolk Associates. Inc. Inc. Inc. Inc. Inc.. Inc. (Japan) Bio-Development SRL Bio-Inova Life Sciences International Biomedia Laboratories SDN BHD bioMérieux.. Germany Dade Behring Inc.Active Membership (as of 1 January 2004) Sustaining Members Abbott Laboratories American Association for Clinical Chemistry Bayer Corporation BD Beckman Coulter.Kalamazoo. Procter & Gamble Pharmaceuticals. Inc. Inc. Ltd. . Inc. Inc.A. Canada) Franciscan Shared Laboratory (WI) Fresno Community Hospital and Medical Center Frye Regional Medical Center (NC) Gambro BCT (CO) Gamma Dynacare Medical Laboratories (ON. of Public Health Laboratories Assoc. L. Inc. John Hospital (TX) Clarian Health . Inc.C. MI Pfizer Italia Srl Powers Consulting Services Premier Inc. SYN X Pharma Inc. Ltd. Instrumentation Laboratory International Technidyne Corporation I-STAT Corporation Johnson and Johnson Pharmaceutical Research and Development. Elanco Animal Health Electa Lab s. Inc. Public Health and the Environment Swedish Institute for Infectious Disease Control Industry Members AB Biodisk Abbott Laboratories Abbott Laboratories. Novartis Pharmaceuticals Corporation Ortho-Clinical Diagnostics.A. CLMA College of American Pathologists GlaxoSmithKline Ortho-Clinical Diagnostics. (MO) Biometrology Consultants Bio-Rad Laboratories. Inc. Ltd.A. Vysis. YD Consultant YD Diagnostics (Seoul. . Beckman Coulter. NY) Ortho-McNeil Pharmaceutical Oxoid Inc. Inc. Hoffman-La Roche AG Fort Dodge Animal Health General Hospital Vienna (Austria) Gen-Probe GenVault GlaxoSmithKline Greiner Bio-One Inc. CT Bayer Medical Ltd.r. Inc. Second Opinion Seraphim Life Sciences Consulting LLC Streck Laboratories. Rhoads Associates.P. Diagnostic Products Corporation Eiken Chemical Company. Clinical Micro Sensors The Clinical Microbiology Institute Cognigen Community Medical Center (NJ) Copan Diagnostics Inc. Inc. Assn.Deerfield.K. Inc. LAB-Interlink. Inc. Inc. IL) Theravance Inc. CA Dade Behring Inc. AP) ARUP at University Hospital (UT) Associated Regional & University Pathologists (UT) Atlantic Health System (NJ) Aurora Consolidated Laboratories (WI) AZ Sint-Jan (Belgium) Azienda Ospedale Di Lecco (Italy) Baxter Regional Medical Center (AR) Bay Medical Center (MI) Baystate Medical Center (MA) Bbaguas Duzen Laboratories (Turkey) BC Biomedical Laboratories (Surrey. Inc.Australia National Society for Histotechnology. Canada) . Primary Care Diagnostics Beckman Coulter K. Asociacion Mexicana de Bioquimica Clinica A. GE (NY) 121st General Hospital (CA) Academisch Ziekenhuis -VUB (Belgium) Acadiana Medical Laboratories. (Tokyo. .Société Canadienne de Science de Laboratoire Médical Canadian Standards Association Clinical Laboratory Management Association COLA College of American Pathologists College of Medical Laboratory Technologists of Ontario College of Physicians and Surgeons of Saskatchewan ESCMID International Council for Standardization in Haematology International Federation of Biomedical Laboratory Science International Federation of Clinical Chemistry Italian Society of Clinical Biochemistry and Clinical Molecular Biology Japan Society of Clinical Chemistry Japanese Committee for Clinical Laboratory Standards Joint Commission on Accreditation of Healthcare Organizations National Academy of Clinical Biochemistry National Association of Testing Authorities . MediSense Products Acrometrix Corporation Advancis Pharmaceutical Corporation Alifax S. Canada) Duke University Medical Center (NC) Dwight David Eisenhower Army Med.West Haven. Inc. Vicuron Pharmaceuticals Inc. Minigrip/Zip-Pak mvi Sciences (MA) Nippon Becton Dickinson Co. British Society for Antimicrobial Chemotherapy Canadian Society for Medical Laboratory Science . Canada) DesPeres Hospital (MO) Detroit Health Department (MI) Diagnósticos da América S/A (Brazil) Diagnósticos da América S/Z (Brazil) Dr.) SARL Laboratoire Carron (France) Sarstedt. Merck & Company. Inc. Enterprise Analysis Corporation EXPERTech Associates. Ammirati Regulatory Consulting A/S ROSCO AstraZeneca Pharmaceuticals Aventis Axis-Shield POC AS Bayer Corporation . Radiometer Medical A/S Replidyne Roche Diagnostics GmbH Roche Diagnostics. Bio-Rad Laboratories. Immunicon Corporation ImmunoSite.l. QSE Consulting Quintiles. HoffmannLa Roche Inc. AB. Inc. Inc.I. bioMérieux. of Health Saskatchewan Health-Provincial Laboratory Scientific Institute of Public Health. (GA) EMH Regional Medical Center (OH) Emory University Hospital (GA) Enzo Clinical Labs (NY) Fairview-University Medical Center (MN) Florida Hospital East Orlando Focus Technologies (CA) Focus Technologies (VA) Foothills Hospital (Calgary.C. OH) Children's Hospital of Wisconsin Children’s Medical Center of Dallas (TX) CHR St. Inc. Inc. Canada) Carilion Consolidated Laboratory (VA) Carolinas Medical Center (NC) Cathay General Hospital (Taiwan) Cavan General Hospital (Ireland) Central Texas Veterans Health Care System Centro Diagnostico Italiano (Milano. Radiometer America. Inc. NY Bayer Corporation . Japan) Medical Industry Association of Australia Associate Active Members 31st Medical Group/SGSL (APO. Roche Laboratories (Div. Machaon Diagnostics Medical Device Consultants. IGEN Inc. Belgium Ministry of Social Affairs. Micro. Cosmetic Ingredient Review Cubist Pharmaceuticals Dade Behring Inc. Professional Members American Academy of Family Physicians American Association for Clinical Chemistry American Association for Respiratory Care American Chemical Society American Medical Technologists American Society for Clinical Laboratory Science American Society for Microbiology American Society of Hematology American Type Culture Collection. IL Dade Behring Inc. Schering Corporation Schleicher & Schuell.L.Glasgow. Tyco Kendall Healthcare Vetoquinol S. Inc. THYMED GmbH The Toledo Hospital (OH) Transasia Engineers Trek Diagnostic Systems. (a Johnson & Johnson Company) LUZ. Canada) Bermuda Hospitals Board Boulder Community Hospital (CO) Brazosport Memorial Hospital (TX) Brooks Air Force Base (TX) Broward General Medical Center (FL) Cadham Provincial Laboratory Calgary Laboratory Services Cape Breton Healthcare Complex (Nova Scotia.Tarrytown. Inc. Italy) Champlain Valley Physicians Hospital (NY) Chang Gung Memorial Hospital (Taiwan) Changi General Hospital (Singapore) Children’s Hospital (NE) Children’s Hospital & Clinics (MN) Children’s Hospital Medical Center (Akron. DE Dade Behring Inc.

Canada) Queen Elizabeth Hospital (Prince Edward Island. Gill. Quebec. M. Canada) Queensland Health Pathology Services (Australia) Quest Diagnostics Incorporated (CA) Quintiles Laboratories.D. Yost. Jones. Canada) Ste. M. Myers. Ph. Canada) Hospital for Sick Children (Toronto. National Institute of Standards and Technology Gary L. John Hospital and Medical Center (MI) St. Luke’s Regional Medical Center (IA) St.S. FAARC TechEd Wayne Brinster BD BOARD OF DIRECTORS Carolyn D..D.D. A Divison of Dianon Systems.. Echevarne (Spain) Laboratório Fleury S/C Ltda. Anthony’s Hospital (FL) St.A. Lincoln Hospital (AZ) John F.. Mary of the Plains Hospital (TX) St. Canada) San Francisco General Hospital (CA) Santa Clara Valley Medical Center (CA) Sentara Williamsburg Community Hospital (VA) Seoul Nat’l University Hospital (Korea) Shands at the University of Florida So.Long Island Jewish Health System Laboratories (NY) North Shore University Hospital (NY) Northwestern Memorial Hospital (IL) Ochsner Clinic Foundation (LA) O.D. Canada) OU Medical Center (OK) Our Lady of the Resurrection Medical Center (IL) Pathology Associates Medical Laboratories (WA) The Permanente Medical Group (CA) Piedmont Hospital (GA) Pocono Medical Center (PA) Presbyterian Hospital of Dallas (TX) Providence Health Care Provincial Laboratory for Public Health (Edmonton. Meyer. Jude Children's Research Hospital (TN) St. Bergen. (GA) Regions Hospital Research Medical Center (MO) Rex Healthcare (NC) Rhode Island Department of Health Laboratories Riverside Medical Center (IL) Riyadh Armed Forces Hospital (Saudi Arabia) Robert Wood Johnson University Hospital (NJ) Royal Columbian Hospital (New Westminster. May. Ph.D. (CA) State of Connecticut Dept.L. BC. Keio University School of Medicine Judith A.D. ON. State of North Carolina Memorial Medical Center (LA) Jefferson Davis Hwy Memorial Medical Center (Jefferson Davis Pkwy. AB. Treasurer The Cleveland Clinic Foundation F.. Joseph’s Hospital . Abbott Laboratories . President CHRISTUS Health Thomas L. Ph. Ph. Vrouwziekenhuis (Belgium) Ordre professionnel des technologists médicaux du Québec Ospedali Riuniti (Italy) The Ottawa Hospital (Ottawa. California Permanente Medical Group South Bend Medical Foundation (IN) South Western Area Pathology Service (Australia) Southern Maine Medical Center Southwest Texas Methodist Hospital (TX) Spartanburg Regional Medical Center (SC) Specialty Laboratories. Ph.D. PQ. M. Davis. Joseph's Hospital & Medical Center (AZ) St. Canada) Winn Army Community Hospital (GA) Winnipeg Regional Health Authority (Winnipeg. President Elect Centers for Disease Control and Prevention Emil Voelkert. M.. Lillian J.D. M. Habig. Ltd. (IL) Jackson Memorial Hospital (FL) Jacobi Medical Center (NY) John C. (CA) St.D. Stephen Kroger. New Orleans. Alan Andersen. Canada) St. Canada) Kimball Medical Center (NJ) King Faisal Specialist Hospital (Saudi Arabia) LabCorp (NC) Laboratoire de Santé Publique du Quebec (Canada) Laboratorio Dr. Secretary Roche Diagnostics GmbH Gerald A. ON. Inc. Canada) Huddinge University Hospital (Sweden) Hunter Area Pathology Service (Australia) Hurley Medical Center (MI) Indiana University Innova Fairfax Hospital (VA) Institute of Medical and Veterinary Science (Australia) International Health Management Associates. Quebec. Centers for Disease Control and Prevention Kiyoaki Watanabe.P.D. Barnabas Medical Center (NJ) St-Eustache Hospital (Quebec. Executive Director Susan Blonshine. Anthony Hospital (CO) St. CAE Canadian Society for Medical Laboratory Science Mary Lou Gantzer. LLC (IN) Middlesex Hospital (CT) Mississippi Baptist Medical Center Montreal Children’s Hospital (Canada) Montreal General Hospital (Canada) National Serology Reference Laboratory (Australia) National University Hospital (Singapore) The Nebraska Medical Center New Britain General Hospital (CT) New England Fertility Institute (CT) New England Medical Center (MA) New Mexico VA Health Care System New York University Medical Center NorDx (ME) North Carolina State Laboratory of Public Health North Central Medical Center (TX) North Shore . Inc. Campus BioMedico (Italy) Loma Linda Mercantile (CA) Louisiana State University Medical Center Lourdes Health System (NJ) Maccabi Medical Care and Health Fund (Israel) Magnolia Regional Health Center (MS) Maimonides Medical Center (NY) Malcolm Grow USAF Medical Center (MD) Marion County Health Department (IN) Martin Luther King/Drew Medical Center (CA) Massachusetts General Hospital (Microbiology Laboratory) MDS Metro Laboratory Services (Burnaby. Canada) Hôpital Maisonneuve . M. Canada) Saad Specialist Hospital (Saudi Arabia) Sahlgrenska Universitetssjukhuset (Sweden) Saint Mary’s Regional Medical Center (NV) St.. Hoeltge. Justine Hospital (Montreal. Canada) Health Network Lab (PA) Health Partners Laboratories (VA) Highlands Regional Medical Center (FL) Hinsdale Hospital (IL) Hoag Memorial Hospital Presbyterian (CA) Holmes Regional Medical Center (FL) Holy Cross Hospital (MD) Hôpital du Sacré-Coeur de Montreal (Montreal.Marshfield Clinic (WI) St. RRT. FCSMLS. Alexius Medical Center (ND) St. Immediate Past President Cosmetic Ingredient Review John V. ON. (Brazil) Laboratory Corporation of America (NJ) LAC and USC Healthcare Network (CA) Lakeland Regional Medical Center (FL) Lancaster General Hospital (PA) LeBonheur Children’s Medical Center (TN) Lewis-Gale Medical Center (VA) L'Hotel-Dieu de Quebec (Canada) Libero Instituto Univ. Canada) Medical College of Virginia Hospital Medicare/Medicaid Certification. LA) Memorial Medical Center (LA) Napoleon Avenue Mercy Hospital (ME) Methodist Hospital (TX) Michigan Department of Community Health Mid America Clinical Laboratories.D. MACP COLA Willie E.. Hearn. M. Canada) Hôpital Saint-Luc (Montreal. Canada) Wishard Memorial Hospital (IN) Yonsei University College of Medicine (Korea) York Hospital (PA) OFFICERS Donna M. RPFT. Ph. Michael’s Hospital (Toronto. Kennedy Medical Center (NJ) Johns Hopkins Medical Institutions (MD) Kadlec Medical Center (WA) Kaiser Permanente (MD) Kangnam St.Rosemont (Montreal. Francis Medical Ctr. J. Ph.D. of Public Health State of Washington Department of Health Stony Brook University Hospital (NY) Stormont-Vail Regional Medical Center (KS) Sun Health-Boswell Hospital (AZ) Sunnybrook Health Science Center (ON. John’s Hospital & Health Center (CA) St.. AdvaMed J.Geisinger Medical Center (PA) General Health System (LA) Grady Memorial Hospital (GA) Guthrie Clinic Laboratories (PA) Hagerstown Medical Laboratory (MD) Hahnemann University Hospital (PA) Harris Methodist Fort Worth (TX) Hartford Hospital (CT) Headwaters Health Authority (Alberta. ON. Inc.. Dade Behring Inc. (OK) UZ-KUL Medical Center (Belgium) VA (Hampton) Medical Center (VA) VA (Hines) Medical Center (IL) VA (Kansas City) Medical Center (MO) VA (San Diego) Medical Center (CA) VA (Tuskegee) Medical Center (AL) Valley Children’s Hospital (CA) Vejle Hospital (Denmark) Virginia Department of Health ViroMed Laboratories (MN) Warren Hospital (NJ) Washington Adventist Hospital (MD) Washoe Medical Center Laboratory (NV) Waterford Regional Hospital (Ireland) Wellstar Health Systems (GA) West Jefferson Medical Center (LA) West Shore Medical Center (MI) Wilford Hall Medical Center (TX) William Beaumont Army Medical Center (TX) William Beaumont Hospital (MI) William Osler Health Centre (Brampton.T. BC.D.(ASCP) Centers for Medicare & Medicaid Services Kurt H. Mary's Hospital (Korea) Kantonsspital (Switzerland) Kenora-Rainy River Regional Laboratory Program (Ontario. Canada) Swedish Medical Center Providence Campus (WA) Temple University Hospital (PA) Tenet Odessa Regional Hospital (TX) The Toledo Hospital (OH) Touro Infirmary (LA) Tripler Army Medical Center (HI) Truman Medical Center (MO) Tuenmun Hospital (Hong Kong) UCLA Medical Center (CA) UCSF Medical Center (CA) UNC Hospitals (NC) Unidad de Patologia Clinica (Mexico) Union Clinical Laboratory (Taiwan) University Hospitals of Cleveland (OH) University of Alabama-Birmingham Hospital University of Chicago Hospitals (IL) University of Colorado Hospital University of Debrecen Medical Health and Science Center (Hungary) University of Illinois Medical Center University of the Ryukyus (Japan) The University of the West Indies University of Virginia Medical Center University of Washington UroCor.. FDA Center for Devices and Radiological Health Robert L. Ph.H. Ph. Canada) Hospital Sousa Martins (Portugal) Hotel Dieu Grace Hospital (Windsor. ON.

0700 T E-MAIL: exoffice@nccls.688.NCCLS T 940 West Valley Road T Suite 1400 T Wayne. PA 19087 T USA T PHONE 610.nccls.org T WEBSITE: www.688.org T ISBN 1-56238-527-5 .0100 FAX 610.

Sign up to vote on this title
UsefulNot useful