Mitigating Risk and Improving Patient Safety – Identify and Eliminate Defects Prior to Stain Review
There has always been pressure on the Histology Lab to do more with less money, less people and less time. Automation in Histology has somewhat stalled over the past few years and is difficult to implement. Automation is only a multidisciplinary process to integrate manufactured machinery and software to perform manual tasks that will allow the lab to capitalize on technology that will provide new and improved processes, increased quality and, most importantly, improved patient safety. The continued evolution of innovation in Histology, both sustaining and disruptive, has taken us from inferior processes to replacement with superior processes. So, why are these processes so critical to success? Developing Innovative Solutions is a demand of the patient and embraced by your management. This workshop will provide learnings/ outcomes from implementation projects and identify opportunities for success and will provide the attendee a new understanding of how automation/ innovation can move your lab from niche application to integrated process.
- Identify opportunities for automation
- Formulate automation/innovation solutions
- Develop improvement processes
Today, we're going to talk about mitigating risk and improving patient safety. There are so many opportunities for us to make improvements, and as we continue to move forward with—we'll summarize it all into precision medicine—it is in our best interest and most importantly the patient's best interest that we move to a better controlled process. The demand from patients, as I've been exposed, is that we no longer provide stains, diagnosis, differential diagnosis but that we now validate our diagnosis and our results. That's very hard for us when we're in a qualitative process where yes and no are our answers.
There's so many different areas when you visualize the process from start to finish in a histology laboratory, but for the purpose of today's discussion what we're going to talk about is the development of a controlled production process. We are producing slides and then stains, and we need to be predictable in what we do, verifiable, and always cost effective. The cost effective part is really pulling at us lately because of continued improvement in the automation opportunities for us in the laboratory. I may have seen two in the world so far laboratories that are completely automated from beginning to end for the histology process. I don't see that as realistic for us in the U.S. any time in the near future, so we are a multi-function process of manual, automated, then back to manual or automated for two steps and back and forth. So all of this creates even more demand on us controlling the production process.
QC of critical steps
We need to get into just the practice of doing multi-point QC for our critical tasks. Accessioning (that's the receipt and labeling of all of our specimens), gross dissection, tissue processing, and the multiple options for staining. Today, we're going to focus on embedding before staining and then staining. So we need to create multi-pole quality control. You know, I still today either as a CAP inspector or a consultant in a laboratory, somebody will run a control slide at the beginning of the day and maybe at the end or wait until the next morning. That's just not going to continue to work for us. So we have this big demand to continue to reduce our variance, and that demand is a compliance part.
But the biggest thing is what I mentioned right at the beginning is we have patients, patient advocacy groups asking us can we validate our results. Yes, we have a pathologist - - and the pathologist is ultimately responsible, but everything and anything the laboratory, histology laboratory can do to reduce that variance will be of value. It's about creating that control process, creating value, and being able to be very, very confident to the validation level of the results that we turn out.
So here’s our responsibility—and here's what we must do. Think of it. In your laboratory today how many of us are using a simple Levey-Jennings chart to track, control, and figure out where we are in our process? Well, it's hard to do when our answers are either yes or no. We need to get to more of a quantitation, and that's going to happen, but we have to start using these simple tools that have been used through other laboratories, GLP, research, and really start to learn what the upper and lower limits of each of our processes are. So that's controlling the process. Defects exist. When we're in a manual process, defects will exist. Now, we have to find a way to identify them, correct them sooner than later, and not allow them to get to the pathologist. But we have to get into a different mindset about what our quality control is.
We also need to understand in that control process where we should put our effort to get the best results, and that's something we're going to discuss a little bit later. You know, when you look at what part of this process am I going to operate, I think one of the most important questions that we should all be asking is what's the cost? What's the effort? What am I going to get out of it? We'll cut through all these other slides, and we'll end this right now by telling tell you that automation can't fix all of our issues, problems, and defects. It still is a reliance and reliability of us to control that process. Until we have a completely automated process from A to Z without any human touch, we have to decide where to put our effort to get the best results.
QC level 1
So in histology, this is how I have approached QC. There are three levels of QC activity that we can attain in any sort of process. The first one is level one. That's where we have so much problems. We're behind in our production. We have conflict between our manual and automation processes, and the technologists have very little time or no time to comply. I think most of us tend to be mired in that level. It's very hard to move to a higher level or a different level when you're constantly looking at how you're just going to produce today's work.
QC level 2
But it is necessary for us to get to level two to where we have enough automation—and automation is necessary because we need walk-away time to put back into our laboratory process quality activities. Now, once we get to level two and you have a certain level of automation—and hopefully some collocation where two pieces of automation are directly or indirectly connected where there is little or no human involvement. Now you create, at the level two, time for the technologist to analyze the process and actually improve the process, not just produce.
QC level 3
Then, our ultimate goal is to get to that level three of quality control. Again, very, very few of us have the opportunity to have a stratification. You need a stratification in your work force where you have bench people, you've got coordinators, technical specialists, supervisors, managers, quality managers. There needs to be a stratification, and let's face it, not all of us have that opportunity, but you're pretty much in a production by automation. That's exactly where the clinical lab is now, and the technologist is constantly engaged in quality assurance. So that's a goal. I think those are realistic goals for us, and I think that's where we have to go in the histology laboratory. It's just that I find that most of us are mired and stuck in the level one.
So what must we do?
We've got to create QC check measurements. Our assessments do not just eliminate the errors. So that's where I get back to having that controlled process. It's very simple. Poor production process will result in poor production results. So you've got to practice a level of inspection at all those various stages in the process. When we're talking about staining we're talking about where are the most opportunities for us to create a defect. That defect is extraneous or as we commonly know them as floaters. Now, there is a lot of information out there about floaters, the incidence of the opportunities to reduce and remove, but, again, think of controlling your process and where you're going to put your effort to get the best results.
So we've got these patient demands, we've got pathologist demands. Slide review is their conversation with the patient, and that slide review starts with the H&E. The H&E is never going away. If they prefer a focused one on one conversation when there is anything else on that tissue section or on that slide that is not anticipated that creates them to slow down, question quality, and that is actually the creation of the defective error. It's a cross-contamination. It typically will create group discussion. Whether that group discussion is with you in the laboratory or with another peer before they come to a decision of what they're going to do. Then, there is compliance and compliance comes from collaboration. It must be a partnership between pathologists, technologists, the process that we've created, and the type of automation that we've placed into the process.
So, again, we go back to how do we validate our tissue diagnosis and we have good verifiable results. Let's focus on the stained slide as it goes to the pathologist. Because I am a CAP inspector and participate in clinic a lot, I tend to look to CAP for direction a lot of times. Their last publication was 2010 with the Q-PROBES investigation. That was retrospective, so they had multiple laboratories go back into all their previous stain slides and find out and prospective. They came up with floaters represent a diagnostic error of about 0.01 to 1.2 % of the slides. Now, that's not acceptable because our goal is always zero, but this is where it starts to get fuzzy. It starts to get complicated whether pathologists all agree that there is a problem, whether laboratories agree, and whether even vendors agree there is a particular problem, but as we move farther and further into the precision medicine process, floaters even at 0.01% can become more of an issue.
So we do have to look at those critical tasks that can contribute to the floaters. Again, Q-PROBES there are designation laws. Gross dissection was a major area of concern, and so was microtomy because of tissue floatation water bath. But all five of those—gross dissection, tissue processing, embedding, microtomy, and then the staining all can contribute to an extra unused tissue piece or floater on a stained and cover slipped slide. Well, I guess there's actually six. It can be us. If we're not wearing gloves all the way through the process there can be cells sloughing off from our touching of slides and that. All of that becomes more and more addition to the discussion as we move to more precise diagnosis. So very simply in-process QC is critical, so you have to find a way to apply your efforts into at least those five areas of our process and make sure that we have a good check. That's Six Sigma. That's Lean. But it's also common sense that you cannot let anything go whether it's a manual step or process *or automated step or process without some sort of continual QC QA. So automation alone can't eliminate it. If you're finding that you have floaters that are not embedded in the tissue, where could they have happened? If they're embedded in the tissue, where could they happen? That's the first step in this, but, again, there are opportunities for defects and errors to happen. Automation can't eliminate it all by itself, and we have to decide where we're going to put our dollars and our effort to get the best results. So what are opportunities?
Opportunities for errors
Gross dissection, there are multiple. When you think about it, someone who—I'm hoping that all of you have had some experience of either participating in the gross room or being oversight or review of that, but multiple specimens moving past an individual. So there are all kinds of opportunity for non-compliance of your quality process. Tissue processing. Putting multiple tissues over into one reaction chamber creates opportunities for defects. Embedding. If we don't have compliance on the cleanliness and how we handle one piece at a time. Opportunity and then microtomy from the cleanliness of the area, the use of any kind of spray refrigerants, the cleaning of the stage, the cleaning of the knife, the cleaning of the water bath, where we place our tissues before and after. All those create opportunity for error. Then staining. If you have an open system or a closed system, what's on that slide or what's in that solution can create opportunity for error.
We wait until the pathologist looks at the slide and says if this is okay or we cannot continue doing so if we're going to be able to have confidence and validate what we do. As you can see, there are multiple areas that create the same result to the pathologist. When the pathologist receives that stained slide, if the suspicious piece is on top or somewhere else on the slide —we use the term floater—I feel that it happened some place other than the grossing dissection bench. If it's embedded into the tissue and it's a clump of cells or at least more than one cell, this becomes more a problem when we get the biopsies. They will then think it's gross dissection or did something happen at collection. Very complicated. Very time consuming to go back and work back through your process, so setting up a good QC QA can give you a lot of time back into the laboratory to continue to work on your production and your quality.
Gross dissection / Fixation
The gross dissection board. There has to be a cleaning each and every time you use specimen forceps. Then, what you're going to do about your fragmented tissue cassette management. You know, is that going to go into a mesh cassette? Is that going to go into some sort of processing paper bag? Something that helps to secure that fragmented tissue. Now, when we have a large piece, even a large dissection and resection piece that goes into a cassette there is an opportunity of sloughing off of cells through the fixation and processing steps. In tissue processing it's all about fixation. If you have poorly fixed tissues, you will tend to have more sloughing off of cells. If you're rushing through your fixation, you may not have a good rinsing of those cassettes to make sure that you removed as much as possible, anything that could be free floating. Then, you've also got your specimen prep and your solution management.
Solution management is very, very important. I still go into laboratories, and people will say well, you know, we change the processor once a week whether it needs it or not. If that's once a day, that's five times that you've exchanged tissues. If you look at the filters in the instruments, cells, multiple cells can easily get into your tissues. We have to set up what is the check and recheck of that process.
Varied opportunities here. Mostly forceps and embedding surface. If you do not have a compliance of cleaning the forceps, cleaning embedding surface in between cassettes, you take the risk of introducing something else into that embedding mold. Then, the third one is always consider the paraffin in the holding area. If you do what they call dry embedding where there is no additional paraffin and it has its own issues, or if you do wet embedding where you have paraffin. Anything that's sitting into the paraffin has a time factor on it, but also because it is molten in a liquid state there is an opportunity for more cells, chunks, pieces, bits to slough off the tissue in the cassette. So, again, a good check, good compliance.
Any time you have under processed tissue you could get pieces of tissue all over the place. If you have properly fixed, properly processed, you should only have—you shouldn't have a fragmenting of your paraffin ribbons and sections. You should have just those sections, and you should be able to keep that whole cutting stage area clean. The more you have fragmentation, you've got poor under processed tissue, the harder it is to keep that clean.
Then, the water bath. What is your method? What is your process? Is it between every ribbon? How do you clean that? How do you identify that? Too often, I still go in and see water baths while people are talking to me and working. There's pieces of tissue floating in there. If you use any sort of aerosol spray, not only do you take the opportunity to introduce that to the water bath, your cutting stage area, but you now—because it's aerosol and you made it airborne, you could introduce floaters and problems, defects into the next cutting station that somebody who is working in a compliant way thinking well, you know, I don't use aerosol. Don't worry about it. Somebody sprays, and now you have tissue fragments that have been airborne. This is one I'm always aware of is slide location and slide storage. Is there more than one slide out in the work area? So all of this can create floaters.
Staining / Drying
We'll talk about the two options we have. We have open container—whether that's a manual or an automated. So we have a certain volume of solution, and then we have the closed system where it puts discrete volume for each step of the process to an individual slide. But in staining the most important thing is making sure that you have the proper slide drying. It works. So drying your slide properly. Because what we're trying to do is not melt off the paraffin. We're trying to get the water. Drain the water from the slide, and out between the paraffin section and that borosilicate piece of glass. The glass is even more important because we have hydrophilic and hydrophobic glass now. They keep changing the chemistry on the glass to where either something is going to stick or not stick, so you have all these different opportunities. All of those create an opportunity for defect, and it doesn't mean whether the manual process, the open automated or the closed is any better or worse for it.
Where do we put our effort and our dollars? Again, before you spend anything for automation or change of process make sure you have a way to check and confirm compliance to the best. Then, solution management—whether it's manual or an automated open system there has to be multi-QC check. Frequency varies. So depending on your volume of slides—if you're in a large laboratory and you're running thousands of slides a day, maybe you're checking every couple of hundred slides. I know everybody is going to pause—well, you mean I got to stop the process, run a QC slide, and before I do anything else? Well, that's the standard method.
You can also do multi-point check of your process by putting QC slides in with a staining rack, but that still means that all your slides need to be reviewed before you put those out, at least the QC slides. If you had nothing on that QC slide, then you're able to put out that rack. Depending on the time constraints on your process, how you can do that multi-point QC check, and then that in process, as we were talking about. The in process is putting a blank slide in a run. Find out how, you know, how many of you out there know that if I put 600 milliliters into a staining vesicle, how many actual slides do I run before I need to change, before I need to replace, rotate? How many slides can you in your process get out of 600 milliliters? The vendors all have the number to throw out there, but what works for you? Have you done a validation and worked with the pathologist to say from slide one to slide X everything looks the same? After that, we either need an addition, change, rotation, clean—clean the container. All of that. Those are the things that too often we just assume I brought an automated instrument. All this works without any further input or effort on our part.
So, again, our number one identifier of all defects—and that happens even in all the laboratories I've worked at—is that yes we run QC. Yes, we need to make sure the process is performing, but the pathologist is still the one that must identify what's under that cover slip. Is it a defect, or is it not and they're going to sign out the case. Then, you have to be able to do some root cause analysis. You know, what is the scope of the issue? Again, somebody comes in and says you need a certain instrument, a certain change in your process because there is a problem with X. Today, we're talking about floaters and extraneous tissue. There needs to be a confirmation for the way you run your process because everybody's step by step process is different. You need to identify what's the scope of the issue, and that takes involvement, partnership and communication with the pathologist to decide is that where—what is the scope and what can be done about it? Often, it may not be that you're going to spend X amount of dollars on a new instrument. It means that you may have to step back and go one, two, or three or four steps into the process and create a new quality assured step by step process.
It does require, though, immediate management. This is not something that you can wait for the pathologist to come to you and say hey, by the way, I think we have a floater issue on our slides. In my experience, that can be anywhere from multiple days to multiple weeks, so it takes immediate management by people in the laboratory. There has to be—if we go back and you think about that first slide with Levey-Jennings, there has to be identification and documentation of all defects and errors. Whether they result in a rejection of a slide by the pathologist, but you need to know and keep daily aware of what the scope and what can be done to do that. Again, expectation of the manual to automated process is that one of the things that I always had angst about is that when we introduced the automated immunohistochemistry instruments. There's one of the most important stains that we do, and also the biggest revenue generator for all of us in the clinical laboratory is that we put this automated instrument at the end of a very manual process. The best instrument out there can have varying performance from lab to lab depending on how well and how controlled your process is prior to placing a slide on that instrument.
It is the same situation when we talk about any automated instrument. Today, we're talking about staining in particular. Before you place that stain, that slide onto that instrument to perform those multi-steps to create your stain whether it's an H&E, a special stain, lot of automation there, or IHC or IF or even further as we go out to all the new techniques that may or may not be used—it is really the responsibility of us to control that process. So what is your expectation for what you're going to get? Then, what the next decision needs to be is am I going to spend that effort and those dollars there, or can I spend the same effort, possibly the least, less amount of effort and less dollars to give me better performance overall in the process. Those are the difficult questions that, that, that come to us each and every day. My worry as a consultant is it's too easy for us to accept that there's an automated solution, and there very well may be one. Depending on what the manual steps are before and after and how well those are controlled, you can almost cancel out the benefits—and you can get to canceling out the benefits of that automated step. So it doesn't matter whether you stay all manual, all automated at this point. It still is incumbent upon us in the histology lab to move to and pass that level one QC and try to get to two and at least one or more people in the laboratory who are functioning at that level three control so that we can get to validating our results.
I'm just going to show you a couple of pictures of what I'm talking about. On the top left, the gross room; That's exactly how clean it should be. You look at the middle. That's how clean the grossing area should be and I will tell you that that's a very rare sight in most of the laboratories I worked in and in most of the laboratories I visit, but that as we move to precision medicine this is exactly the expectation.
Over on the right, these are some of the things I pulled off of different vendor sites. My suggestion to everybody is in the gross room, in embedding, and at microtomy do not use serrated forceps. Look at the opportunity for tissue to be captured into those serrated areas. That can easily be carried over to another piece of tissue and into the next case if there is not 100% compliance of cleaning and maintaining a clean work area each and every time you work with a specimen. Already I can hear people say, you know, we could never get our work done if we did this. Well, yeah we could. Think about the time that we're wasting by backing up and investigating and everything else. Yes, once you set up a good control process you will be able to maintain your productivity. Just like anything else, it's going to take a change, but you will be able to maintain and possibly increase your productivity. Then, it also makes it easier when you have a controlled process to identify, select, and implement the next piece of automation in your laboratory.
Cassettes: I have a lot of issues with cassettes. Why do we decide to force all of the cassettes we can, and if we're leaving that to the person who loads the cassettes—whatever the recommended number by the vendor is, someone will try to put one more cassette per rack. You've got all these different tissues, all these different combinations of complex tissues with an opportunity for something to slough off and move forward. The biggest thing I have with this is that when you use an instrument to capacity—and I can hear all the vendors cringing right now—use the instrument to capacity you're just not going to get the same performance. If you took two or three cassettes out of each one of those rows, you have a much better flow of solutions. You have a better exchange of those solutions.
Then, off of that, if you look at those different cassettes, we go to mesh. You go to different size holes. There are tissue processing papers. I didn't put up my bigger angst, which are sponges, because that is just such a problem with carryover in my mind, but there's so many different options. It really should be one specimen container whether it's fragmented small biopsies or whether it's a resection, one piece of tissue, one set of fragments on cassette. We've got to maintain—if you're not maintaining that or you're trying to put too much into a cassette because you don't want to do a second cassette, just more opportunities for error. Controlling the process also means you have to control everything about the process, even what's going into the cassettes.
The tissue processor. Again, there needs to be a good validation of the frequency chosen for solution changes—if you're running cassette racks such as what is pictured, is once a week the right number? Who knows. You have to validate, so you have to have solution management. There are many instruments out there, and this is where automation comes back in, that have solution management options for you that help you maintain a certain specific gravity or dilution factor in your solutions that are going to be used in processing. What they can't manage at this point is whether or not there's any other tissue floating in those solutions. That is still our responsibility. Your frequency, reducing load capacity, and then making sure your samples are secured. How often do you come into an issue where a cassette may be open and somebody says, you know, what are we going to do about that? Or you pull out the rack and there are tissue fragments in the bottom. Throw away those tissue fragments. Unless you do DNA sampling and you know who those patients were, there's no way of identifying who they belong to. That's an indicator that you may have to do something different in your cassette management if you're seeing fragments of tissue left in the chamber. Well, cassette management or fixation, grossing technique, we have multiple opportunities here to make sure this changes.
Embedding. Again, these are some things right off of the vendor sites. Multiple specimens sitting open around. Making sure that you've got non-serrated forceps. Things like that. It is—this is one of those areas along with microtomy where working with one cassette at a time and making sure there's a compliant cleaning in between is helpful.
Microtomy. There's so many. Again, just some pictures out there. On the bottom left, if you got good process, good fixed tissue, that should be what you're looking at while you're trimming and everything else. We have in the middle we have some trim just leaving all of that debris. All it takes is somebody to cough or whatever. You've got tissue floaters. The work area needs to be clean in the upper, the middle—on the upper. There's tissue fragments laying all over the counter. Lay a slide down. It flips over. You now have tissue fragments. Then, my favorite is—this was right off of somebody's YouTube video of how to do microtomy and pick up tissues on a tissue flotation bath, and I didn't even circle the fragment that is sitting right next to the top ribbon, but there's tissue fragments all over the water bath. It is incumbent upon us as technicians and technologists to work in the highest craft that we can. That is not a good craft controlled process there. Those are where I see the opportunities for floaters and extraneous tissues occur way before I get to a stainer. Now, if I control all of those I have a much easier time controlling the staining (whether manual, open automated, or closed automated process) and being able to validate and have confidence in that stained cover slipped slide.
So, again, solution management. In process QC checks. Frequent solution change. Then, make sure that you have completely drained and dry slides before you move into your staining process. All of those, you concentrate on that. You have a better opportunity to be successful with your automation. Immunohistochemistry, there are so many different things, but fixation techniques. It creates a cross-link with proteins. We got all these antigen retrievals. You have incompletely processed tissue samples that we're trying to do immunohistochemistry on. We need to pay more attention before we ever get to cutting a block for immunohistochemistry.
Again, what are the improvement opportunities? Those are the five, in my mind, that we have to do a good level and get past level one QC, but often I find that there's not even a good level one QC process that's multi-point check of identifying, correcting as soon as possible. Again, back to the other premise I had is you should hopefully be able to get to three different levels of QC in your laboratory. I understand the difficulty with the constraints and the pressure on producing and why it keeps us in the level one, but you have to find each—every lab you have to find a way to get the first person to start doing level two and the first person to start doing level three or you're continually working and reworking your results. You may ultimately spend very valuable and not so frequently available dollars in the wrong place if you can't get to those other levels of QC.
So control the process. Whether it's a manual, automated, open, closed, it requires rigorous QC by us. That's time consuming, but it does require it. You've got to be able to create effective and multi-point QC checks. Better QC gives you cost effective automation. Automation can't fix your errors and defects alone. You have to be in a controlled process and participating and complying before you can really gain the benefits of whatever that automation is you're going to purchase.
As I mentioned, here are some the references. I encourage all of you to look at this and see what's out there. It's very hard to get consensus across all the laboratories and pathologists that there's a big problem. I think it's important that we start addressing this now before we move to any further precision in our staining techniques.
What would be the mean time in days to change a battery of H&E stain in a frozen section room performing around ten procedures a day to ensure a good stain on slides?
That's pretty thoughtful. Because we have no standards in histology, that would be a validation protocol that you should do. In my world your number of frozen sections is a very low volume laboratory compared to some of mine doing 60 plus cases a day. There is no set standard, so it's hard to tell. There should be a filtering of solutions on a regular basis. Depending on the type of tissues that you're working with, I would suggest that if you're in doing all diagnostic biopsies or all skin that your frequency of filtering solutions would probably be a little higher if you're doing a resection piece of tumors. But all of that depending on how fast you're freezing, how well you're able to cut, you have a very high opportunity that you may have a fragmentation of tissue that could come off in your solutions. So, I would say some sort of multiple time through the day you would want to filter and/or change, whichever is most cost effective. Then, doing it by validation would tell you exactly how often you might have fragmented tissue that's sloughing off of your frozen section slide.
What are the three main areas of concern for creating floaters?
Okay, so I have my own personal preference, but I'll go back to the CAP Q-PROBES. I would say the number one is your microtomy area just because anything that has not happened correctly in your process, fixation, processing and everything, all lands to the person who is creating the slides at microtomy once they put that block into the chuck. That has to be one piece, and cleanliness has to be complied with. The next biggest area is the gross room. Any drift in compliance of clean instruments and a clean grossing area whether you're using pads or rinsing in between using a hard surface, that is the second one for me. Third would be embedding. Again, if you follow good operating procedure, standard operating procedures, and comply, you should be able to reduce that way below the point down to that 0.01. Remember that in a manual process we can never remove all of the defects. Maybe on a day for a time period, but, again, microtomy, then grossing, and then embedding, but they demand a one piece flow and then a cleanliness compliance in between those units or pieces.
During staining what is your QC recommendation for validating a product, for example, a detection kit which constitutes various subcomponents obtained from multiple vendors?
Well, I think, first of all, I would go to your CAP guidelines. They give you a lot of good information about how you're going to do lot to lot validation. Second, pathologist always needs to be involved, and you have to then first design your plan of what type of staining you're going to do when you develop your protocol. That's not unusual for a laboratory to have an antibody detection kit and everything from multiple vendors. There are opportunities, but I don't think most labs are to the point where they're buying everything from one vendor. It is just—and there's a lot of information out there between CAP and other resources of how you develop that validation protocol. Statistics come into here. One slide is not a validation. Two or three is probably not a validation, so you have to have that conversation of where you're going to get to a statistical number that says if I look at all of these and one's out and 5 are in, is that a correct stain or not? In my world, statistics start at about six samples. That's probably the very lowest N, but that is a discussion between guidelines, regulations, and pathologists—medical director or pathologist over your area deciding what is acceptable for you to do that. Any time you change any one of those components, a new validation has to happen.
Can you explain the three levels of QC you referenced in your presentation?
Level one is we're just trying to maintain our high production, and you have put in place check, identify, and correct at least at all of your critical steps. That takes some time, effort, and participation. Level two is now you're moving to more of a quality assurance because you're now with the level one I've also stated is probably your biggest data collection area. You're collecting the data. Level two is now you're working with the capabilities and capacity of automation and/or with the data working towards quality assurance steps to continue to reduce your defects and errors. Level three, you're working with all of the what ifs. Now that I have all of this, if I do this what will change? Level three is almost all quality assurance. You still have a certain number of people doing level one, a smaller group doing two, and a smaller group of three. The difficulty—maybe I didn't explain it well—is that too many times I walk into a lab and even on certain days in our lab everyone is involved in level one, and you're never going to get past just identifying defects without being able to get to those other two levels and start working on ways to prevent the opportunity before it even happens.
Do you recommend emptying the paraffin tray on the microtome between cutting each block?
Oh, that's a tough question. There are multiple opportunities there. Cleanliness is most important. I've seen a couple things. People remove and dump their tray. I have also been—and it was in a different country—when they have their collection area for their trimmings before they go to cut the block once they—so they'll clean their stage—their microtome knife stage area, and then they will place a clean paper towel over their trimmings before they cut their next ribbon. So it is either a frequent dumping or something such as I saw at a different in Europe where they place a paper towel over all of their trimmings before they start cutting their next specimen ribbon.