Q and A

December 2007

Q. Can you explain why we are required to perform extensive lot change studies for prothrombin time and partial thromboplastin time? These include PT reference range confirmation for the International Normalized Ratio (INR) formula, activated partial thromboplastin time VIII and IX sensitivities, heparin sensitivities, correlation of multiple instruments, and control range calculation, and yet point-of-care instrument vendors are able to supply calibrated reagent strips or cuvettes. This is a very expensive and time—consuming annual activity. Why can't lab coagulation suppliers manufacture a calibrated reagent?

A. Coagulation systems report results for a variety of purposes, such as to monitor anticoagulant therapy, screen for functional defects or deficiencies, and diagnose coagulopathies. While some analytes are best measured using chromogenic and immunoturbidimetric methods, many functional assays use a "time to clot"-based methodology. Differences between reagents and testing systems are known to contribute to result variability, making crossover studies necessary when using new reagents or implementing new testing systems.

All central lab coagulation systems use some form of clotting-based methodology, but point-of-care PT devices may use clotting time or electrochemical, fluorescence, or impedance outputs. These values must be correlated to a time unit if the output is to be interpreted as a comparison to that from the central lab system. Because of differences in reagent presentation, the results obtained by those devices that use clotting times do not correspond with the results obtained by central lab systems. Because the output of each manufacturer's device is unique and is reported in units unfamiliar to the central lab, it is necessary for the manufacturer to translate this information into a usable format for the customer. To overcome this hurdle, POC devices use bar codes or magnetic strips, which often carry calibration data specific to the lot of cartridge being used, to calculate the final result.

Besides calibration data, POC device manufacturers need to provide everything necessary to report an interpretable result for a sample. In the case of the PT/INR, this includes a reagent International Sensitivity Index, the mean normal time for the local population, and the patient test result. The manufacturer must provide values for what is considered a population-specific variable, such as mean normal time, because the task is too onerous, complicated, and expensive for customer labs to perform, even if they were provided with proprietary information and algorithms.

A similar scenario unfolds for the APTT. In POC devices, APTT assays are not carried out as traditional two-stage assays, with a preliminary two- to three-minute contact activation step, followed by the addition of calcium chloride as a starter reagent. For one APTT POC device, the APTT reagent is premixed with calcium and the sample reconstitutes the reagent, then the process is monitored for clotting time. In a normal sample, the clotting time may be 70 to 80 seconds, which is converted through a slope and offset calculation to 30 to 32 seconds.

While POC devices fulfill a necessary function in coagulation testing, ease of use and convenience come at a price. The costs include a lack of correspondence with central lab results, inability to conveniently or cost-effectively provide and use local population information, and reduced precision compared with central lab analyzers. Traditional coagulation testing will continue to be used as the measuring standard because of significant advantages in precision, flexibility, throughput, and cost.

Mark X. Triscott, PhD
Vice President, R&D
Reagent Development
Instrumentation Laboratory
Orangeburg, NY

Q. Where would I find information that I can use to set up new guidelines for peripheral smear review and urinalysis reflex testing to culture?

A. I discuss when to review a blood smear in a question and answer in the January 2006 issue of CAP TODAY, page 85. The Q&A makes reference to an article written on behalf of the CAP Hematology and Clinical Microscopy Resource Committee (Lab Hematol. 2001;7:175-179), which you may find helpful.

Urine testing that should lead to reflex to culture is a trickier issue. An American Academy of Pediatrics guideline (Pediatrics. 1999; 103: 843) says if you want to rule out a urinary tract infection in a child you must do a culture, and that no urinalysis test is sufficiently sensitive. The guideline provides a lot of data on the value of various dipstick tests. In adult females, treatment is often undertaken without a culture and is based on symptoms and urinalysis. For adult urinalysis, I think you need to arrive at a procedure that represents the consensus of your clinicians. I do not think that laboratory reflexing to culture will be acceptable to clinicians and payers.