Printer Friendly

A troubleshooting guide to quality control.

IN THESE DAYS of regulation it has grown increasingly important to perform well on proficiency testing events. CLIA is yet another reason for top-notch quality control.

Poor QC inevitably leads to poor PT--and a possible test procedure failure from HCFA surveyors. The penalty is serious: loss of the right to perform the failed test until proficiency has been demonstrated on a designated number of consecutive PT examinations.

I derived these suggestions from my experience with chemistry controls at our large group practice laboratory. Much is applicable to other lab sections. Figure 1 lists common control problems and their typical causes. Figure 2 lists ideas for preventing and solving QC problems.

* Cost. While the amount of money spent on a control is important, it should not be the only factor.

Choose the appropriate number of levels of controls--at least two unless only one is available. Controls should monitor normal/therapeutic and elevated/toxic (callback) patient ranges. If using three control levels, select abnormally low or trough patient ranges.

* Assayed or not? It is usually cheaper to utilize unassayed controls, assuming that the test will be performed frequently enough to establish in-house control ranges.

Unassayed controls. Run comparison studies between your current and unassayed controls. Perform at least 20 separate runs, ideally over 20 days or more. Ask the vendor how many other laboratories use the same unassayed controls. The larger the number, the greater your opportunity for comparison--and for increasing your confidence in your in-house ranges.

Assayed controls. Before purchasing assayed controls, review manufacturers' inserts. If only a few of your methods are included and the ranges are too wide for your purposes, choose something else.

Ask the vendor how many laboratories and reagent lot numbers were used to establish the assayed ranges; the more labs and numbers used, the more accurate the range. One advantage of using assayed controls is that their predetermined ranges will make it unnecessary to conduct very many comparison studies with the controls you were using before. As a general rule, I conduct five comparison studies with assayed and at least 20 studies with unassayed controls. Of course, CLIA requires evaluation whether or not controls are assayed.

* Constituents. Using controls that contain a sufficient number of constituents reduces the number of kinds of controls that are needed, thus saving time and money. Even so, certain special constituents, such as acid phosphatase and ammonia, must often be obtained as single-constituent controls.

* State. Should you choose freeze-dried or liquid controls?

Lyophilized controls must be reconstituted. The stability for most constituents other than enzymes tends to be good. Lyophilized controls are usually stored at refrigerator temperature whether they have been reconstituted or not.

Liquid controls do not require reconstitution. The stability of every constituent, including enzymes, is usually good. Liquid controls are generally stored at freezer temperature until opened and at refrigerator temperature after being opened. Their ease of use makes them a good choice for satellite sites and other locations where instruments are operated by nonlaboratorians. Disadvantages include high cost and the inability to use some liquid controls with dry chemistry methods.

* Size. Does the vendor offer a bottle size that suits your needs? If the control bottle is too small, you will be required to reconstitute the control frequently. Overlarge bottles encourage waste.

* Frequency of use. When purchasing for tests that require special controls and are performed infrequently, select controls that can be aliquoted and frozen to maintain stability.


* Sequestering. Sequestering a large supply of a single lot is cost-effective and assists in troubleshooting activities. I try to buy enough of each lot number to last for an extended period. The range must be established only once, while control lot numbers have to be correlated less often. Further, having an ample supply of one lot facilitates the monitoring of control shifts or trends. Gathering multiple control values over time leads to a well-established mean and range.

It's important to check with the vendor to make sure that the QC program will run as long as you plan to keep the control. Evaluate the new before the old expires. Contact the vendor if a problem is found.

* Different sources. Use daily controls from a manufacturer other than that of the instrument or reagent undergoing QC--but keep the latter on hand for troubleshooting.

* Outside sources. When choosing controls, find out if the manufacturer or vendor offers a comparative QC program. Such a program is useful in monitoring shifts and trends and for troubleshooting instrument and reagent problems.

Ask whether the manufacturer will send you a summary of your results if you submit statistics or graphs by computer, mail, or fax. If so, what form will the reports take? How often will they be generated? Will they be cumulative for the month or year? Will all data submitted by the lab be included, or will outliers be excluded? What statistical information (mean, standard deviation, number of labs, number of values) will be incorporated? Will you receive a breakdown of means and SDs by type of instrument and reagent?

* Attend to basics. By choosing appropriate controls and monitoring them successfully, we help to optimize instrument performance and enhance the accuracy of test results. Expert QC also aids the staff in performing well on PT events.

Figure 1

Reasons underlying some common control problems

Shift within acceptable range

Lyophilized control

Improper mixing

Left at room temperature too long

Bottle-to-bottle variation, especially if reconstitution procedure


Liquid control

Improper mixing

Bottle-to-bottle variation within reagent lot number

Change in reagent lot number (typical of enzymes; ordinarily

produces slight shift only) Shift outside acceptable range Any of the causes listed above Concentration of control (in some tests, altered by low volume of

lyophilized or liquid controls in the container--e.g., 0.5 ml in

a 5.0-ml bottle) Control improperly reconstituted Reagent contaminated (can occur within new or used lot number) Specialty controls (e.g., bilirubin) exposed to light too long Instrument problem (sampling, reagent delivery, mixing)

Trending (results moving away from mean) Change in instrument's reaction temperature Gradually worsening instrument sampling problem Flawed delivery of reagent by instrument

Excessive scattering (results spread out across range) Improper mixing in instrument Contamination during testing

Figure 2

What to do when controls are off the mean

To resolve problems such as those listed in Figure 1, take the simplest steps first. Some guidelines follow.

1. Check the dates on which the control was opened and will expire. This should have been indicated on the bottle.

2. Check the dates on which the reagent was opened and will expire.

3. Mix the control properly; rerun. (Most lyophilized controls are mixed by gentle inversion, most liquid controls by swirling.) This often works, especially when the problem lies with one level of control.

4. If your laboratory participates in a QC program, examine the summarized reports to look for any shifts in the mean. If your laboratory is not in a QC program, ask the manufacturer of the control whether any other labs have reported problems with it recently.

5. Mix the reagent; rerun the control.

6. If a lyophilized control is the problem, reconstitute a new bottle with fresh diluent. Use a volumetric pipet, not a serologic one. Follow the manufacturer's instructions carefully.

7. If a liquid control is the problem, open and use a new bottle.

8. Recalibrate the reagent, especially if two or more controls have shifted.

9. If the controls shift after a new reagent lot number has been started, rerun the normal and abnormal patient specimens that were run with the control bearing the old lot number. If patient correlations are good, the control shifts are probably acceptable. If they are poor, the reagent may be bad.

10. Try a new lot number of the reagent. If that corrects the problem, ask the reagent manufacturer to find out whether any problems have been reported with the old one. You may perform a public service by alerting the manufacturer to a reagent problem that other labs may encounter.

11. Troubleshoot the instrument. Many kinds of malfunction can disturb the controls. Check sampling, reagent delivery, mixing, lamp integrity, and reaction temperature.
COPYRIGHT 1993 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Johnson, Judy E.
Publication:Medical Laboratory Observer
Date:Sep 1, 1993
Previous Article:Strategies for handling imposed changes.
Next Article:Computer technology of the not-too-distant future.

Related Articles
Tailoring clinical chemistry instruction to the job market; a survey of area labs pinpointed needed changes in this MT curriculum.
How to find particleboard defects.
Course covers rubber extrusion.
Sun Chemical introduces innovative troubleshooting guides. (Fresh Ink).
Remote diagnostics monitors vital signs in the lab.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters