As the director of laboratory operations of the Department of Pathology and Laboratory Medicine at UCLA Health, Paul Colonna oversees eight labs. Collectively, the labs process millions of specimens a year, most of which are patient blood, urine, and other bodily fluids.
These specimens are organized and catalogued through a laboratory information system (LIS), which has a tracking function that can give the lab managers and those involved a clear map of where the samples are located.
“That’s what we have right now, but that’s a very manual system, which means that if I want [a particular] specimen, I have to send someone to go get it,” Colonna says. “They have to look at the computer, find out where the spot is, find out which refrigerator it’s in, what shelf it’s on, and what number is on the rack [in order] to pull it out.”
The somewhat manual system, while to a certain degree useful, is not the most efficient, especially with so many samples and refrigerators scattered throughout the different locations.
A better system would be one that could not only retrieve a sample but could do so in a more systematic way that would aid those involved in research. An example would be a setup that could categorize and pull out specimens based on parameters such as age and gender.
Such a system would prove useful if, for instance, researchers wanted to pull out samples from all females over 40 whose calcium levels were low, Colonna explains. “I want to be able to pull that information and I want the system to go through all the specimens in that refrigerator. I want it to pull any information or any parameter out that I want.”
But such a system doesn’t currently exist. The closest to what Colonna and many in his position are seeking is a Roche Cobas p701 Post Analytical System—also known as the Stockyard—a refrigerator unit attached to an automated line that can store specimens. It’s interfaced with a LIS, which will allow a healthcare professional to type in a name and have the system fetch the correct specimen, bring it out, run the test, and place it back into storage. Finally, at a specified date range—such as in three or five days—the system will throw specimens away.
While the Stockyard is an improvement over what the UCLA labs have, it’s not a smart enough system to justify its cost.
“I would get that system, and certainly it would help to some extent, but looking at its capabilities, it really is very limited, so [do] I want to spend $800,000 for a smart refrigerator that is not real[ly] smart?” he says. “In other words, this is the 21st century. We can send spaceships to Mars; we should be able to pull different specimens out of a refrigerator.”
However, Tim Streit, IT product manager for Roche, says that with some modification the Stockyard could be used for specific parameters. In the case of the calcium example that Colonna brought up, he explains that “in conjunction with Roche Middleware Solutions [an IT solution that allows for this ability], a workspace could be created where they could have columns with a category, [such as] ‘Age >40,’ and apply a filter for this and also another column, [such as] ‘Low CA,’ and apply a filter to it based on whatever is considered low.”
“This,” Streit says, “would then show you all of the samples that would meet that category. You can then manually request the instrument to retrieve them. Middleware is used today for the retrieval of samples from the p501/701 by looking up the specimen; it just requires a bit of tweaking.”
From collection to storage
Though sophisticated out-of-the-box systems are currently not available, laboratories such as the ones at UCLA each have to come up with a scheme that makes storage and retrieval as simple and accurate as possible with what’s on the market and suitable for their work.
“Having a place for proper storage is critical, and that starts with the collection of the sample,” says John Consolvo, who manages the environmental laboratories at the Philadelphia Water Department’s Central Laboratory. “Many samples have a temperature preservation requirement. Those samples need to be placed in a cooler, on ice, immediately upon collection for transport back to the lab.”
Each month, Consolvo and his team analyze approximately 3,000 water and wastewater regulatory and research samples for a suite of biological and chemical parameters. Each sample type has a specific holding time requirement. Some tests are performed right in the field, immediately upon collection. But most are returned to the laboratory for analysis.
Once received and signed for by the appropriate receiving lab, the sample may need to be stored in a refrigerator, freezer, or incubator if analysis won’t begin immediately.
“So not only is having storage necessary, but procedures for checking and documenting that the sample was maintained at an appropriate temperature throughout storage are also necessary,” he says. “Even if samples can be maintained at room temperature prior to beginning the analytical processing, having a place to store the samples out of the way of the daily workflow, away from potential sources of contamination, is very important.”
When a sample is collected and brought into their central receiving unit (CRU), technicians sign for it, and may temporarily store the sample in a refrigerator in the CRU while they are logging the requested test(s) into a laboratory information management system (LIMS). They then notify the appropriate labs, and a lab analyst comes to the CRU to sign for it and either store the sample or prepare it for analysis.
“We have areas within each lab’s refrigerators and freezers assigned to hold specific types of samples, or samples assigned for specific testing,” Consolvo says.
Diana Crary, manager at Ronald Reagan Hospital’s CORE lab, and who also works for Colonna at UCLA Health, agrees that a well-streamlined system that begins during sample collection is an important aspect of storage.
“With our new EPIC Beaker Laboratory Information System, a lot of our samples already come in labeled with bar codes, and they are also designated by tube color, so everything is matched to that sample,” she says. “So if I get two red tops, they will have two unique identifiers with the same patient on there. We scan and receive them in, and then they will pass up to the different labs where they need to go.”
At UCLA Health, each specimen is assigned a bar code. Let’s say a phlebotomist draws a patient’s blood into tubes. Each tube would have a separate unique identifier. If there are, say, five tubes associated with one person, each one of the tube bar codes will have a .1, .2, .3, .4, or .5—each one of the dots has a test associated with it. This helps keep tests organized and easy to catalog.
But not everyone finds the need to use such a system. Consolvo’s lab, for instance, does not—in their sample collection—utilize bar coding—partly because as a water utility laboratory, most of the lab’s customers are internal to the utility, so evaluating this technology has not been a high priority.
“Our laboratory’s CRU prepares the sample bottles. This includes labeling the bottles,” he says. “Some bottles are very specific to the test(s) to be performed, so these labels can be prefilled with the test(s) to be performed. There are lines on the label for the sample collector to fill out the date, time, and location of sample collection. This information is then immediately transferred to a chain of custody at the time of collection.”
Despite the system in place, Consolvo knows that as his facility becomes increasingly paperless, they might eventually need to consider bar coding.
“Many of our more advanced instrumentation has been software-driven for a long time, but we’re now slowly implementing software- based bench sheets and logs for some of our more general wet chemistry bench tests,” he says. “I think that we will eventually look for ways to apply this paperless concept even to sample collection, and bar coding seems like a direction labs are moving toward, particularly medical-based testing labs.”
For most labs, the main requirement for specimen storage is temperature. Each facility decides what temperature they need to set their refrigerators at, depending on the specimen.
“Temperature requirements vary depending on what the sample is to be tested for,” Consolvo says. “Most of our samples require storage temperatures to be around 4°C. These samples are placed in a cooler on ice immediately upon collection, and then transferred to a refrigerator upon receipt in the lab if they won’t be analyzed immediately.”
To store their samples, Consolvo’s lab staff use both plastic and glass bottles, depending on what each sample is being tested for: samples that will be used to test for organic chemistry parameters are collected in glass containers.
“There is information in each analysis method’s standard operating procedure (SOP) about the type and volume of container, as well as preservative (if any) and whether the preservative is in the bottle at the time of collection, or added after sample collection, back in the lab,” he explains.
Colonna’s labs, however, only use plastic tubes, mainly because it’s safer in terms of potential breakage and causing harm to those handling them. The other advantage of this container type, according to Crary, is that plastic allows samples that include gel separators to be frozen.
The team keeps samples anywhere between one day and three months: “We store things for different lengths of time, depending on what we have room for, because again, we’re at different locations,” says Colonna.
Being able to store samples is especially useful if a physician wants to add an additional test. Instead of going back and drawing more blood from a patient, lab professionals might be able to use available specimens if they are still viable.
“So our laboratory automation system is smart enough to know what tests have been ordered, what tubes have been drawn, and if the specimen is still viable to use for a variety of different tests,” Colonna says.
Despite the capabilities of the current system, Colonna says that having a truly smart and automated storage system that’s ready to go would save him and others many hours searching for specimens with specific characteristics to be used for validation and research: “To be honest with you, if a company developed something like that, I promise you I’d buy it tomorrow,” he says. “I don’t care what it costs. Because the reality is, the benefits I’ll get out of it are huge.”