Monday, April 4, 2011
COAGULATION CORNER: APRIL 2011
Quality Control and Medical Decision Levels in Coagulation
QC ensures better patient care by assuring the accuracy and precision of laboratory resting. Having acceptable QC results validates the reliability of instruments and reagents as well as allowing result to be reported with confidence.
So what do we want from our QC? We want it to detect an error in the coagulation process- the reagents, the analyzer or the operator.
This error can result in a clinical error that may result in a change in diagnosis or treating a patient. For example if a PT result is 12.0 sec versus a 12.5 sec is that a clinical error? No- but it may be an analytical error. It is important for laboratories to understand what their coefficient of variation (CV) for their testing at several different levels in particular at medical decision points. A medical decision point is defined as the concentration of analyte being analyzed, at which some medical action is indicated for proper patient care. There may be several medical decision levels for a given analyte. In coagulation the best way to determine medical decision levels by looking at where intervention may be required, and also to understand at what CV a test that answer becomes relevant versus just being inherent variability within a test result.
CLIA includes criteria for acceptable performance for three coagulation tests in its hematology category:
So what does that tell me? I have acceptable limits that tell me how my test is performing, but then I need to understand medical decisions- remember the PT at 12.0 versus 12.5? Well what happens when that result goes from 12.0 to 13.0? Does that mean that something is happening in my patient? Do I need to watch to see if medication needs to be adjusted, or is that just the variation in my test? How was my QC, was it within limits, but on the high side? Lots of questions here- so how do I answer them?
First- QC- we know that 2 levels should be run, we want a normal level, and what about the second level? There are 2 ways to choose that level- it can be a high level, in particular if your patients are on anticoagulation, and you want to make sure that your reagent works well in that range, or where your total system (analyzer and reagents) are challenged- where product may have to be given, where a patient may bleed. Those levels should be determined by the laboratory director to best reflect your patient population. Okay so step one is we have good control over how our reagents perform.
Second- what is a significant CV? This is such an important question- it helps to understand how your test performs, so how do you know what this is? The best way to demonstrate this is to perform precision testing- at a minimum to look at inter-laboratory variability. Precision is the reproducibility of the result when repeatedly measured in the test system. Running at least 10 repeats ( or 20) at a low, normal and high level, can help to determine if a change in a result is inherent in the testing system, or really a significant change in result. Lets look at a Protein C result in the low range- of 30%, and the clinician wants to know if the level is improving- since the second result is 35%? From precision testing you know that at 30%, the CV is 20%, that means result of 24-36% are probably the same result, and not directly related to an increase in result. That is a lot of information to be able to provide regarding results.
So what about controls on the high side? We know that the performance of analytical methods can be monitored by analyzing specimens whose concentration are known and by comparing them to the observed values with known levels. So what do we expect from our controls? We expect stabile controls with little vial to vial variation and to have the same test matrix as our specimens. We also assume that each control has the same chance of detecting an error, which is not always a valid assumption. Error is some assays occur more readily at higher concentrations than at lower in particular in calibration curves are not linear. It is also important to be aware of reagent degradation and contamination. And no matter how many times it is said, your source of water and pipettes are still an important pre-analytical variable.
So while QC is necessary in the laboratory and can be a painful and time consuming task, we know it is important is how a result is used, and how the physician makes a decision.
About the Author
This website contains information on products which is targeted to a wide range of audiences and could contain product details or information otherwise not accessible or valid in your country. Please be aware that we do not take any responsibility for accessing such information which may not comply with any legal process, regulation, registration or usage in the country of your origin.