College of American Pathologists
Printable Version

  Queries and Comments


May 2005
Originally published in CAP TODAY

Stephen J. Sarewitz, MD

From the files of the CAP's checklist-related questions. Answers reviewed by Stephen J. Sarewitz, MD, LAP checklist commissioner and staff pathologist, Valley Medical Center, Renton, Wash.

Q: With regard to analytic measurement ranges for blood gas and co-oximetry (Nova Biomedical), we are having difficulty finding a material to validate our reportable ranges. PCO2 only goes to 80 mmhg; pO2 only validates to the mid-300s; and the co-oximetry material does not validate the normal carboxyhemoglobin and methemoglobin parameters. Using this material means we will have AMR violations on a daily basis. Since we can't find a material to fit our needs, can we extrapolate these results to extend our ranges (not to exceed the manufacturer's specifications) to cover the routine values we see? For example, carboxyhemoglobin material validates the range from 4.3 to greater than 20. The normal range for carboxyhemoglobin is approximately zero to two. We can validate the abnormal results but not the normal results.

A: In the case of blood gas analysis, the requirements for analytic measurement range verification apply only to those parameters that can be calibrated and that are measured directly. These requirements do not apply to calculated parameters. Most oximeters can be calibrated for only total hemoglobin. Therefore, AMR verification should be performed for total hemoglobin, but it is not needed for derived quantities, such as carboxyhemoglobin fraction or methemoglobin fraction. Similarly, AMR verification is appropriate for pH, pCO2, and pO2, but not for calculated bicarbonate or total CO2 concentration.

The materials used for AMR validation must be known to have matrix characteristics appropriate for the method. The test specimens must have analyte values that, at a minimum, are near the low, midpoint, and high values of the AMR. Guidelines for analyte levels near the low and high range of the AMR should be determined by the laboratory director. Factors to consider are the expected analytic imprecision near the limits, clinical impact of errors near the limits, and availability of test specimens near the limits.

It may be difficult to obtain specimens with values near the limits for some analytes—for example, T-uptake, free thyroxine, free phenytoin, prolactin, follicle-stimulating hormone, troponin, and pO2. In such cases, one should adopt reasonable procedures based on available materials. The method manufacturer's instructions for validating the AMR should be followed when available.

Q: Checklist item CHM.30550 requires the analytical measurement range for sweat chlorides to be no more than 10 mmol/L. Our instrument directly reads down to only 50 mmol/L. Is this a problem?

A: Checklist item CHM.30550 requires the lower level of the analytic measurement range to be less than or equal to 10 mmol/L. Most individuals without cystic fibrosis have sweat chlorides of less than 10 mmol/L. Getting an exact number is important for sweat chloride results in the 10 mmol/L to 45 mmol/L range because such results are abnormal and require clinical followup even though individuals with chlorides in this range do not have cystic fibrosis.

Q: Do checklist items TRM. 31241 and TRM.31375 apply to blood filters and administration sets even if a department other than the laboratory is responsible for distributing and using these materials? For documenting the inspection and testing of materials like this, would it be sufficient to indicate that the correct stock number was received? For ongoing tracking, would it be sufficient to document the receipt of a shipment of a certain lot number and then document the use of that lot number per patient? Furthermore, how long would these records need to be retained?

A: A critical material is considered a good or a supply used in the collection, preservation, storage, testing, or transfusion of blood components that directly affect quality or patient safety. Some examples of critical materials include reagents used for pretransfusion testing or filters and administration sets issued by the laboratory for blood transfusion.

The expectation for TRM.31375 is that your laboratory would have an inventory control system in place to track the use of all lot numbers of critical materials received, which would include date received, placed into use, discarded, or returned to the supplier. The purpose of this requirement is to ensure that your laboratory can determine which lot number was used for any given patient. If your laboratory determines that there was a problem with a lot number of a critical material or is notified of a defect in a critical material, you would be able to determine which patients may be affected. Some laboratories are using paper-based systems—for example, inventory control log sheet—for tracking. I have also seen computerized spreadsheets used for this purpose.

For situations where inventory and use of critical materials bypasses the lab, the lab should have an agreement with those areas outlining the responsibility for inventory control. In addition to indicating that a correct stock number was received, a physical inspection of materials should be performed, as well as testing, if applicable, to document acceptance of receipt of new lots (TRM.31241).

The College requires that records relating to quality control be retained for five years.

Q: Do I need to perform parallel testing for waived point-of-care tests, such as urine pregnancy and dipsticks? If I am using the same type of materials in the laboratory urinalysis area, do I need to do parallel testing for general lab purposes?

A: Verification of reagent performance is required and must be documented for all areas of the laboratory, including POC testing sites. Any of several methods may be appropriate, such as direct analysis with reference materials, parallel testing of old versus new reagents, and checking against routine controls. Checklist items POC.05000 and POC.05020 indicate that new reagents must be checked using an appropriate method and the results recorded before being placed in service.

Where individually packaged reagents or kits are used, criteria should be established for monitoring reagent quality and stability based on volume of use and storage requirements. Periodic processing of wet controls to validate reagent quality and operator technique is a typical component of such a system. If possible, one should use previously tested patient specimens for reagent lot comparisons.