Return to CAP Home
Printable Version

  Q & A





cap today

March 2001

Q.  What is the requirement for quality control on sedimentation rates?

In standardizing hematology procedures across our laboratory network, we have discovered a variety of ways that labs are performing QC on ESRs. Some are purchasing controls and run a normal and abnormal control every shift (and spending $4,000 per year). I think that is a little overboard, especially considering they never reject a run because it is "out of control." Other labs have developed various methods, such as having two technologists perform the test on the same sample periodically, to ensure control. We document the room temperature and that the plate is level. All of our labs are participating in CAP proficiency testing, but I don’t necessarily feel that since QC material is now available for ESR you have to purchase and run it. This is a test we’d like to offer at more of our clinic locations, but purchasing QC material would drive up the cost and probably force us to migrate the testing to the core labs.

Our labs are using various methods with ESRs, so perhaps we need to look at what the manufacturers suggest and follow their guidance (but they sell QC material—of course they are going to suggest you run it).

What is the requirement for manual retic QC? We have made QC slides that are low, normal, and high, and they are available to the technologist at the microscope. But that does not control the staining process. Should we purchase retic controls and take them through the whole procedure?

A. From a CAP accreditation perspective, the essential elements are that all QC programs make scientific sense, there is documentation of the program structure, records are kept, and appropriate actions are taken in response to QC data. Additionally, we have the constraints imposed by federal law.

Inasmuch as manual ESRs are classified as waived tests under CLIA ’88, there are no federal regulatory specifications for QC of this test. I personally do not believe that ESRs should be viewed as having the same QC requirements as a complete blood count or other truly quantitative hematology tests. Rather, the ESR should be viewed as a semiquantitative test, even though the results are numeric. From a QC perspective, manual ESRs are more akin to a qualitative test, so that two-level QC material is excessive. However, some semiautomated and automated methods have been classified as moderately complex under CLIA ’88 and may therefore require two-level QC.

In the CAP Laboratory Accreditation Program hematology checklist, we have no specifically defined QC for ESRs. However, a Laboratory General checklist question ("If the laboratory performs test procedures for which calibration and control materials are not available, have procedures been established to verify the reliability of patient test results?") is applicable to ESRs, since there are no calibrators for ESR tests, and the question refers to calibration and control materials. In other words, it is up to the director to define how ESRs will be controlled, and this can be accomplished in a variety of ways that do not require purchased materials. Extramural proficiency testing, attention to details of specimen collection, prompt setup, direct observation of test performance, and patient replicates by different technologists are entirely appropriate.

I agree with your assessment that the commercial availability of liquid controls for ESRs does not mean laboratories should be compelled to use them, particularly if one is trying to contain costs. Failure to reject a patient run because the commercial QC material is out of control is a clear violation of the law and sound laboratory principles. Further, it is a wasted expense, inasmuch as the QC results are not guiding decision-making about release of patient results.

QC for manual retics is a different situation than manual ESRs. Under CLIA ’88, this is a moderate-complexity quantitative hematology test [Fed Register. 1993;58 (141):39931] for which the federally mandated QC requirements at 42CFR493.1202(c)(4) do apply. In other words, at least two levels of controls with known QC ranges must be employed on each day of testing. Liquid commercial control materials are convenient, because one takes them through the entire process of slide preparation, staining, and microscopic evaluation. Of course, there is no guarantee that matrix effects with some preparations could produce different staining characteristics than fresh patient blood. Your use of prepared slides of authentic patient specimens to assess inter-technologist imprecision is entirely appropriate as a QC activity, although technically you need only two-not three-levels for regulatory purposes.

Albert Rabinovitch, MD, PhD
Chief medical officer/vice president
Specialty Laboratories
Santa Monica, Calif.

Checklist Commissioner
CAP Commission on
Laboratory Accreditation

Q.  The CAP requires curve checks for factor assays performed on automated analyzers when using the same lot number of deficient substrate as curve stored in the analyzer. What are the guidelines to know how closely the curve check values must agree with the stored value?

A. Good laboratory practice, as well as regulations, requires that any assay, including coagulation factor assays, be calibrated over the relevant range and that controls be performed regularly to confirm assay calibration. At least two controls must be performed; however, if both reduced and elevated levels of a given analyte are of clinical importance, controls should be performed at the clinical relevant levels of the assay, both above and below the reference range.

More specifically, coagulation factor assays performed on automated analyzers can have their calibration curve stored in the instrument’s computers and used for up to six months after they are originally determined. Repeat calibration is required with any change made in the maintenance of the instrument or a change in any of the reagents involved in the method.

Each time a coagulation factor assay is performed, the validation of the curve is accomplished by the use of controls. The acceptance/rejection criteria of a given assay based on these controls are best determined in the laboratory. The within- and between-run precision is measured by performing multiple determinations of the aliquots of the control and calculating the mean and standard deviation of values generated.

The controls should have concentrations of the factor in question within the reference range and below the reference range. (Some use a control at the mild deficiency level of approximately 20 percent and a more severe deficiency level of less than 10 percent to validate the curve.) In the case of factors VIII, IX, and VII, as more data continue to appear on the importance of elevated levels of these factors in the evaluation of thrombophilia and potential thrombotic complications, calibration and controls with high concentrations of factor are also indicated.

The method for determining the acceptable range for controls for coagulation factors is not different than what you would do for any analyte in the laboratory. Begin the process by performing a minimum of 20 determinations with the control to determine the mean and standard deviations for the assay. Following this, you can set up the standard Levy-Jennings quality control chart for each one of the controls being used. In general, coagulation factor assays are quite precise, generating CV of approximately five to 10 percent when the control is at 1 unit/mL (100 percent). At lower concentrations, the CV will rise. Once the limits are developed and the Levy-Jennings charts are in place, the standard multirule approach to determining if the assay is in control can be used, just as it would be used for other chemical analytes.

Suggested Reading
1. Westgard JO, et al. Basic QC Principles. Madison, Wis.: Westgard Quality Corp. 1998.
2. Arkin CF, Bovill EG, Brandt JT, et al. Factors affecting the performance of factor VIII coagulant assays. Results of proficiency surveys for the College of American Pathologists. Arch Pathol Lab Med. 1992;116(9): 908-915.

John D. Olson, MD, PhD
Professor and director
of Clinical Laboratories
Department of Pathology
University of Texas
Health Science Center
San Antonio
Chair, Coagulation
Resource Committee

Q.  Our pathology department often gets specimens such as synovial or breast aspirations. The specimens are brought to the department in syringes that have been recapped. We would like to have these specimens brought to us in sterile tubes or containers. Some physicians seem to feel that important cells could be lost by removing the needle. Are there any guidelines to follow regarding collection and transporting of pathology liquids? Is there any way to avoid recapping and hand carrying a syringe through the facility hallways?

A. The following recommendations are from the National Committee for Clinical Laboratory Standards. These are well-known manuals that offer recommendations to pathology laboratories on various subjects.

First, the material from solid lesions that are aspirated through a needle should be expressed onto a glass slide immediately after the material is obtained. Otherwise, clotting can occur and block the needle hub. The needle can be rinsed in any transport medium to collect the residual material within the needle. Transport media that are frequently used include cell-culture media, balanced electrolyte solution, or denatured ethyl alcohol solution. Saline and formalin are not recommended. The transport medium can be centrifuged or filtered. Cytolyt or Cytorich, which is required with the new preparations devices (ThinPrep and AutoCyte Pap smears), can also be used for transportation.

For liquids obtained by fine needle aspiration, NCCLS has a specific manual that addresses in detail the best method for collecting and preparing various fluids, including synovial and breast cysts. This manual even describes how specimens should be identified (labeled) and transported and includes rejection criteria. This manual is called NonGynecologic Cytology Specimens: Collection and Cytopreparatory Techniques (Vol. 17, No. 19).

These recommendations, if you decide to use them, should be incorporated into your policy and procedure manual and enforced. The clinicians need to be informed of these policies, and inservices given to them or appropriate personnel regarding these changes in the procedures. Stress that "under no circumstances is it appropriate to submit a needle to a laboratory for processing." This statement is from the NCCLS Fine Needle Aspiration Technique Guidelines (Vol. 16,No. 7).

The NCCLS documents can be purchased. To order by phone, call the NCCLStoll-free number, 877-447-1888.

Janet F. Stastny, DO
Johnson City, Tenn.

Dina Mody, MD
Baylor College of Medicine

Dr. Stastny is a member of the Cytopathology Committee.

Dr. Mody is vice chair of the Cytopathology Committee