Skip to main content

Quest Diagnostics Teams With Copan to Expand Automation of Clinical Microbiology

Premium

NEW YORK – With the recent launch of an automated microbiology system in its Marlborough, Massachusetts laboratory, Quest Diagnostics is looking to streamline a department that has seen less impact from automation than many other parts of the lab.

Sourced from lab automation firm Copan Diagnostics, the system is anticipated to cut costs and turnaround time and help address challenges facing clinical microbiology labs, including staffing shortages and reimbursement cuts.

While automation has made significant inroads in lab disciplines like clinical chemistry and hematology, microbiology still relies heavily on manual processes.

"There hasn't been a lot of change in the field over the last 30 to 40 years," said Matthew Mackechnie, microbiology laboratory supervisor at the Marlborough lab, noting that microbiology workflows typically require many sample touchpoints, which has made automating them difficult.

First samples have to be sorted and batched according to what kind of test will be run on them. Technicians then culture specimens on media plates that have to be handled and sorted, then incubated, and then sorted again based on what amount of growth is observed.

"On a typical low-complexity sample, you're going to handle something six, seven, eight times before a result goes out to a client," Mackechnie said. "What automation has done is reduce that to one or two times."

One of the major challenges to automating microbiology work is dealing with the diversity of sample containers used, he said. "Hematology, chemistry, they have standard tubes that they use. There are only four or five tubes, whereas in microbiology we have 40 or 50 different sample types we could see in a day, all in different [containers], so it's really challenging to create a machine that can handle all that."

To address this, the new system installed at Quest uses liquid transport media, which allows for use of standardized containers. Urine samples also come in standardized containers. Altogether, around 85 percent of the samples come standardized in this way, Mackechnie said, noting that these samples can be loaded directly into the system, which sorts and plates them automatically.

For negative samples, a technician won't have to touch the sample between loading it into the system and reporting the results to the client, he said, adding that "that is really amazing for microbiology."

Positive samples require more handling, but the process here has been streamlined, as well. After plating, samples are delivered on a track system to incubators where they are imaged at intervals to look for culture growth. Whereas technicians once had to sort through incubating plates by hand to assess whether they were showing growth, they can now just check the images, which is a significantly faster process. The system also includes software that uses artificial intelligence to analyze the images for signs of culture growth before the technician takes a look, allowing them to complete their analysis more quickly.

In addition to reducing manual sample handling and sorting, the imaging system helps address what Michael Mitchell, director of microbiology services at the University of Massachusetts Memorial Medical Center, which is served by the Quest lab, said was sometimes a mismatch between the timing of staff shifts and the incubation process.

Cultures are typically checked in a batch once per shift, and "different cultures need to be incubated for certain amounts of time," he said. "Depending on when they go in [the incubator] that may not correlate with [staff] shifts. If something goes in late, it may not have incubated long enough when the morning shift [checks it] so it may have to wait another eight or even 24 hours."

Images, on the other hand, can be pulled singly throughout a shift when they are ready to be examined, "which really improves efficiency," Mitchell said.

Another advantage is that because technicians aren't manually sorting through incubating plates, they aren't disturbing the incubation environment, which leads to more consistent conditions and faster growth.

"In traditional micro, you're in and out of the incubators all the time," Mitchell said. "So things cool down to room temperature, then it has to go up again. The CO2 levels might go down." He added that at 16 hours of incubation in the automated system, plates often show as much growth as they would from a full day in a traditional incubator.

This could prove particularly valuable for antibiotic susceptibility testing where time is of the essence, Mackechnie said.

He noted that he and his colleagues were still evaluating what all the potential advantages of the system might be.

"The first thing, obviously, is to implement it and then to find out where the gains are," he said. "We're seeing that growth rates are so much better. So, now what can we do? How will that let us get results out faster?"

Mitchell said the lab hasn't done a formal evaluation of the cost or time savings provided by the system but said it is clear from his experience, thus far, that it takes fewer people to process more samples.

Mackechnie said the lab was now working to implement a component it called the "collaborative robot" that would allow it to automate handling of the roughly 15 percent of samples that don't arrive in a standardized container.

"For instance, if we get a tissue sample from someone in the [operating room], we will actually have to take that sample and do a bit of manipulation of it before we can create the culture," he said. "What the robot will do is… you'll have the sample in hand and the robot will present you with the labeled media, and all you will have to do is cut the sample and do that first inoculation of the media and the robot will take it from there."

"What that does is create one standard process for the technologists," he said. "They are looking at one plate using the same inventory no matter what the sample type is."

The lab plans to roll out this technology in the next month or two, Mackechnie said.

Microbiology is a ripe target for automation not only because it has retained a large amount of manual work but also because of a confluence of rising sample volumes, an increased emphasis on more rapidly identifying infections and antibiotic resistances, and a declining workforce.

Staffing is a challenge for almost all parts of the clinical lab, but microbiology is among the most impacted. In part this is due to increased centralization of microbiology labs, which has reduced the number of slots available for clinical rotations.

Increasing automation will likely hasten this trend toward centralization even as it helps address the lack of manpower.

"To get to this level of integrated automation is expensive," Mitchell said. "Quest is in a position to leverage its size to be able to provide the efficiencies from this, and not all small labs can do that."

Large laboratory firms like Quest have been active in acquiring the operations of hospital labs in recent years and it believes automated systems like that implemented at the Marlborough facility could help it win more microbiology business. The company provided an analysis that projected that "a midsize hospital that transitions their microbiology operations to Quest can experience as much as $5 million in cost savings associated with lower equipment costs, labor, reduced hospital stay days and other factors."

At the same time, microbiology departments are relatively well suited to weather the reimbursement cuts driven by the Protecting Access to Medicare Act that have played a major role in many hospitals' decisions to offload their lab businesses.

Amanda Schmidt, senior marketing manager for Copan, noted that because the company's systems are modular, smaller hospital labs without the resources of Quest can automate at least portions of their workflow.

"A lab can start out with the upfront specimen processing [unit] and then add modules as they go," she said. "And that makes for an exciting proposition for, say, a hospital lab that doesn't have the kind of money Quest has, for example."

Silvio Lignarolo, senior business development manager at Copan, noted that the company has seen substantial interest in its systems from smaller hospitals.

"For Quest it is a way to benefit by increasing their volumes," he said. "For smaller hospitals, automation is, I think, a chance for them to remain relevant and retain the business that they have by limiting the impact of lower reimbursements."