Dr Alexander von Meyer
EFLM Preanalytical Working Group Chair
Dr. Alexander von Meyer, EFLM Preanalytical Working Group Chair, answers your questions put forward during Bio-Rad’s webinar on QC strategy for Serum Indices (HIL) detection, covering topics such as: the value of pre-analytical focus, best-practice strategies for improving quality assurance, and what to do if your HIL interference checks fail.
Disclaimer: The views, information, or opinions expressed in this document are obtained from and approved by Dr. Alexander von Meyer and from publications in European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) website and do not necessarily represent those of Bio-Rad Laboratories and its employees. The primary purpose of this Q&A is for information only.
|Q1:||Is there a good reference you can send to participants that explains the HIL indices?|
|Dr von Meyer: More background can be found at the EFLM website.|
|Q2:||Can we perform the biochemistry parameter in hemolysis sample? Or is it better to reject the sample? Background: in dialysis patients some samples we are receiving are hemolyzed.|
|Dr von Meyer: More background can be found in the following two papers:
1. Managing hemolyzed samples in clinical laboratories. Simundic AM, Baird G, Cadamuro J, Costelloe SJ, Lippi G.Crit Rev Clin Lab Sci. 2020 Jan;57(1):1-21. doi: 10.1080/10408363.2019.1664391. Epub 2019 Oct 11.
2. Practical recommendations for managing hemolyzed samples in clinical chemistry testing. Lippi G, Cadamuro J, von Meyer A, Simundic AM; European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) Working Group for Preanalytical Phase (WG-PRE). Clin Chem Lab Med. 2018 Apr 25;56(5):718-727. doi: 10.1515/cclm-2017-1104.
|Q3:||Is the EFLM providing manufacturers with guidelines for quantitative cut-off values for the HIL interferences? So that the manufacturers of the serum indices material as well as IVD manufacturers could use these?|
|Dr von Meyer: No. Cut-off values are always test/assay dependent. Therefore, they cannot be provided by a scientific organization. The EFLM want to improve the way and the amount manufacturer can give information about their assays and the interference on these assays to customers. More data are available or can be available what is important for the laboratory to provide higher quality. That is the actual approach of the Task and Finish Group of Prof. Simundic.|
|Q4:||When you have a serum sample with strong lipemia, what do you do? Probably you discard it. But if that sample is an emergency sample what do you do? Is there a correct procedure to manage it?|
|Dr von Meyer: More background can be found in the following papers:
1. Lipemia Index and screening for hyperlipidemia - A diagnostic opportunity?. Gils C, Thorsen AF, Nybo M. Clin Chim Acta. 2020 Feb;501:83-84. doi: 10.1016/j.cca.2019.10.024. Epub 2019 Nov 26. PMID:31784091.
2. Lipemia: causes, interference mechanisms, detection and management. Nikolac N.Biochem Med (Zagreb). 2014 Feb 15;24(1):57-67. doi: 10.11613/BM.2014.008. eCollection 2014. Review.PMID: 24627715.
|Q5:||Could automatic HIL detection replace visual check? If yes, how can you be sure that each HIL result answered by the analyzer is accurate?|
|Dr von Meyer: It is for sure better than any visual check. In TLA you will not have a chance to do a manual check. Evaluation and QC provides the stable high quality of HIL testing.
1. Glick MR, Ryder KW, Glick SJ, Woods JR. Unreliable visual estimation of the incidence and amount of turbidity, hemolysis, and icterus in serum from hospitalized patients. Clin Chem 1989;35:837-9.
2. Lippi G, Cadamuro J. Visual assessment of sample quality: Quo usque tandem? Clin Chem Lab Med 2018;56:513-5.
3. Luksic AH, Nikolac Gabaj N, Miler M, Dukic L, Bakliza A, Simundic AM. Visual assessment of hemolysis affects patient safety. Clin Chem Lab Med 2018;56:574-81.
|Q6:||Is it mandatory or necessary to do automatic HIL detection if we previously made a visual check in the pre-analytical part?|
|Dr von Meyer: Sometimes even very low concentrations of interferences can cause clinically relevant deviation of the “real” value. These low concentrations cannot be determined visually. Many publications show the superiority of the automated determination of HIL indices over the human eye.
Such undetected interference can cause harm to the patient. Therefore, from our point of view visual determination of interferences is questionable concerning overall quality of lab services. We showed this aspect in many publications to convince laboratory professionals as well as manufactures to rely only on automated HIL assessment.
|Q7:||Do you (or the pre-analytical WG) have any advice on what cut-off values you use to determine whether to reject a sample or not? Do you use for instance cut-offs based on the reference change value (including biological variation) or do you use the analytical change limit based on analytical variation only?|
|Dr von Meyer: Please see article Managing hemolyzed samples in clinical laboratories. Simundic AM, Baird G, Cadamuro J, Costelloe SJ, Lippi G.Crit Rev Clin Lab Sci. 2020 Jan;57(1):1-21. doi: 10.1080/10408363.2019.1664391. Epub 2019 Oct 11.|
|Q8:||Is there any strategy/advice with other interferences such as certain drugs or pathologies?|
|Dr von Meyer: First of all, it is necessary to get this info from clinicians. This is often difficult and not reliable. If this information is available a comment on the report is recommended. Because we mostly do not know the concentration of the “interferent” it is difficult to say if an interference is taking place in a specific sample. But also, here more studies are welcomed.|
|Q9:||Do you have any recommendations to interpret HIL values for icterus and lipemia? For these cases, a pre-analytical phase error such as hemolysis is not common.|
|Dr von Meyer: You mention a complex topic. You are right, that you redraw a new sample is not changing the result in most cases. Therefore, two approaches can be considered: 1. With an interferogram you can set-up a second cut-off for reporting the values after discussion with the clinicians for specific patient groups e.g. ICU patients with icterus. 2. Establishing and evaluation of approaches to reduce the interferent in a sample. For some analytes the results are ok as a compromise.|
|Q10:||How do you recommend to handle a grossly icteric sample in a patient with severe liver disease? You will not be able to obtain a better sample, as you would with a redraw of a hemolyzed sample.|
|Dr von Meyer: That is a well described problem. Here we would recommend an interferogram to get an idea of how much a deviation from a "real" value could be. In these patients I know from discussion with clinicians a very broad range of allowed deviation could be possible. This is online with the CLSI recommendation where explicit clinical expertise is mentioned for setting up a cut-off. With an interferogram you can have the discussions with the clinicians and setup a second cutoff for reporting these values to a specific group of patients.|
|Q11:||Does hemolysis affect the results?|
|Dr von Meyer: Yes, sometimes even very low concentrations of interferences can cause clinically relevant deviation of the “real” value.|
|Q12:||Is HIL grade compensating for a real biochemical reaction: a real quantification for the analyte?|
|Dr von Meyer: HIL just offers a possibility to get an idea if an interference is occurring. It is not so precise that a compensation is recommendable. There is literature on this topic. Some colleagues tried compensation e.g. for hemolysis. But there are reasons not doing it, so we do not recommend any compensation approaches. A comment to the report for resampling is recommended.|
|Q13:||How do you improve potassium if a patient is not in a condition to collect another sample?|
|Dr von Meyer: I guess the question is related to hemolysis and potassium. The first thing to consider is the type of hemolysis: in-vitro vs in-vivo. in the case of in-vivo hemolysis the potassium value can be considered as real. In the case of in-vitro hemolysis resampling is recommended as the optimum. If it is not possible it is valuable to have an interferogram to get an idea of how big the deviation from the real value can be expected. This can lead to a twostep reporting system (see literature).|
Learn why pre-analytical focus is crucial to ensuring quality results in your laboratory. Watch a short webinar for more insights from Dr von Meyer. Register for free here.
|Q1:||Should QC follow up be numerical or categorical (semi-qualitative)? Our analyzer only gives the category, and the manufacturer does not give ‘raw’ absorbance information.|
|Dr von Meyer: This is a known problem. We as the WG-PRE would like to change this, as I mentioned. Please ask your manufacturer if he can provide it? Tell him why it is necessary. Together we could probably change this situation to deliver higher quality for the patient.|
|Q2:||Should we report HIL indices to clinicians quantitatively? If yes, in what units?|
|Dr von Meyer: It depends on your clinicians. We do not do it. It would be too complicated for them. I you use the two-step approach, which I mentioned (paper by Lippi et al.) you would have a middle way. Without own validation you cannot give the results with units to the clinicians because they are just indices, so without units. You can do something like a validation for an inhouse method based on these indices, e.g. for hemolysis. Then you have an LDT with the responsibility in your lab. This can improve for example cases in transfusion medicine.|
|Q3:||Could you repeat which would be the frequency of testing HIL QC?|
|Dr von Meyer: In line with your routine test intervals. We would recommend to handle this QC comparable to other QCs. This makes it also easier to implement.|
|Q4:||What are precision and accuracy recommendations for HIL indices?|
|Dr von Meyer: The first step will be to harmonize reporting, after all manufacturers report index in a continuous scale, precision and accuracy recommendations will be considered.|
|Q5:||How many levels should the HIL QC have?|
|Dr von Meyer: At least two.|
|Q6:||What QC limits would you suggest on HIL QC measurements. Classic Westgard rules like 2SD and 3 SD or rather a 6-sigma approach?|
|Dr von Meyer: Please see Professor Dr Lippi's paper; Local quality assurance of serum or plasma (HIL) indices. Lippi G, Cadamuro J, von Meyer A, Simundic AM, on behalf of the EFLM Working Group for Preanalytical Phase (WG-PRE) Clin Biochem 2018;54:112-8.|
|Q7:||If QC for HIL is out of range, but all analyte QC was acceptable, what is the recommended strategy for auto-validation of patient results? Do we hold all samples that should have HIL?|
|Dr von Meyer: This seems it makes HIL QC possible to have a very big impact for any false flags. Theoretically - HIL QC range is much wider… Yes block the system if HIL is out of range…|
|Q8:||Would you apply Westgard rules for HIL QC?|
|Dr von Meyer: Please see Professor Dr Lippi's paper; Local quality assurance of serum or plasma (HIL) indices. Lippi G, Cadamuro J, von Meyer A, Simundic AM, on behalf of the EFLM Working Group for Preanalytical Phase (WG-PRE) Clin Biochem 2018;54:112-8|
Find out how to create a robust quality assurance strategy for pre-analytical work in your laboratory. Join a short webinar for more insights from Dr von Meyer. Learn more.
|Q1:||How can we detect hemolysis invisible to the human eye?|
|Dr von Meyer: Sometimes even very low concentrations of interferences can cause clinically relevant deviation of the “real” value. These low concentrations cannot be determined visually. Many publications show the superiority of the automated determination of HIL indices over the human eye. Such undetected interference can cause harm to the patient.|
Therefore, from our point of view visual determination of interferences is questionable concerning overall quality of lab services. We showed this aspect in many publications to convince laboratory professionals as well as manufactures to rely only on automated HIL assessment.
|Q2:||There is not a good understanding that the L index is typically a measure of light scatter/turbidity rather than the typical visual lipemia evaluation that is more closely related to triglycerides concentration.|
|Dr von Meyer: You are right. Literature "Heterogeneity of manufacturers' declarations for lipemia interference — An urgent call for standardization" show the problem you mentioned.|
|Q3:||In some places, laboratories have a pre-analytical instrument that is different from the test analyzer, in which one should I run the HIL QC?|
|Dr von Meyer: On clin-chem instrument, where the check is done - where there is a photometer.|
|Q4:||How is HIL reported on the analyzer?|
|Dr von Meyer: It depends on the analyzer. Some manufacturer reports already continuous values (as long as I know Roche and Abbott). Others report only semi-quantitative values, which cannot be used in the same way like continuous results to optimize a report in favor for the patient. So, we would recommend if you cannot report continuous values today, ask your manufacturer for it. There of us ask for it, the more likely we will get it.|
|Q5:||How does one go about validating the HIL indices in the laboratory (e.g. when a new platform is validated)? What is your experience with comparability of HIL indices among different vendor platforms (how similar/different they are)?|
|Dr von Meyer: Validating can be done in relation to literature or standards (e.g. Lippi et al; Multicenter evaluation of the hemolysis index in automated clinical chemistry systems). The comparability of hemolysis and icterus is pretty much comparable if continuous scales are used, because both are pretty much related to the free hemoglobin or the bilirubin concentration. For lipemia this is not the case. Comparability between two systems using qualitative or semi quantitative results are difficult.|
|Q6:||What do you think of the new approach for some manufacturers to detect HIL interference by camera instead of testing in analyzer with saline?|
|Dr von Meyer: From current info this is not comparable and not sensitive enough - could be valuable for pre-check.|
|Q7:||You mention that a lot of instruments report 0, +, ++ or +++. and that this is not ideal. What other way to show the HIL analysis do you recommend instead?|
|Dr von Meyer: Continuous results, like 15,34 for hemolysis. It should be absolutely comparable to all other results of the analyzers. From a technical point of view the analyzers are measuring precise results, but the instrument software blur it to a qualitative result.|
|Q8:||How can you process the lipemia sample in wet chemistry? In bilirubin test indirect test values are giving negative values so is any lypoclear solution available? Or is there any other method to perform the test?|
|Dr von Meyer: There are solutions like lipoclear on the market. Also approaches which high centrifugation is shown in literature. But this all needs to be evaluated by every laboratory if there is no manufacturer recommendation.|
|Q9:||How often did you pick up an analytical problem with HIL doing QC?|
|Bio-Rad: There is no formal study to support the frequency of analytical problems with the HIL module. We don’t have data to support this question.|
|Q9:||Are there established ranges for the HILs?|
|Bio-Rad: Per CLSI C56AE, the range is shown below:|
|Bio-Rad Liquichek Serum Indices is designed to target the following concentration on the Ortho VITROS 3600:|
|Note: Recovery may differ depending on the make and model of the instrument.|
Learn how to protect against QC failure during the HIL detection process. Gain more insights from Dr von Meyer in this short webinar. Click here to join.
|Q1:||Does Bio-Rad offer QC for Indices?|
|Bio-Rad: Bio-Rad offers Liquichek Serum Indices intended for use as part of laboratory interference testing to monitor an instrument’s response in detecting hemolyzed, icteric, and lipemic (HIL) samples.
The Bio-Rad product line has separate kits containing Hemolysis, Icterus, or Lipemia targeting at common chemistry assays HIL interference claim. Bio-Rad provides a Non-Interfered kit that acts as a negative sample for customer’s convenience.
|Q2:||What kind of QC material is acceptable for HIL indices?|
|Bio-Rad: QC material containing human sourced material prepared following procedures similar to what was established in the CLSI guideline C56-A and EP07.|
|Q3:||Do you have a good peer group with the Liquichek Serum Indices, using Roche Cobas Integra and 6000?|
|Bio-Rad: Bio-Rad's Serum Indices product can be reported in their Unity QC Data Management system and currently has over one thousand data points for the Roche Cobas 6000 / 8000 / c 311 instruments in our Unity Peer Group.|
Join Dr von Meyer to learn about recommendations from the EFLM (European Federation of Clinical Chemistry and Laboratory Medicine) working group on pre-analytical evaluation. Register for our free webinar.
Dr Alexander von Meyer
EFLM Preanalytical Working Group Chair
Dr von Meyer earned his Medical Degree from the University of Munich and also holds an MBA from the University of Applied Science in Neu-Ulm. He is Director of two Institutes for Laboratory Medicine providing Lab services for 5 hospitals in the north of Bavaria in Germany. Dr von Meyer is the Head of the German Working Group “extra-analytical quality” of the DGKL and is also a member of two of the EFLM Working Group on “Preanaytical Phase” and on “Postanalytical Phase”.