Laboratory Acquired Infections

Transcript

Sabrina DeBose, DHSc, MS, RBP

Centers for Disease Control and Prevention

DLSbiosafety@cdc.gov

Didactic Speaker

Sheldon Campbell, MD, PhD, FCAP

Professor of Laboratory Medicine, Yale School of Medicine

Associate Chief for Laboratory Medicine, VA Connecticut Health Care

sheldon.campbell@yale.edu

David Peaper, MD, PhD, D(ABMM)

Associate Professor of Laboratory Medicine, Yale School of Medicine

Director, Clinical Microbiology Laboratory, Yale-New Haven Hospital

david.peaper@yale.edu

Aufra Araujo: Alright. Good afternoon, good morning, and good evening, everyone. My name is Aufra Araujo, and I want to extend a warm welcome from the Centers for Disease Control and Prevention in Atlanta, Georgia. I am a PhD health scientist in CDC's Division of Laboratory Systems, and I've been facilitating these ECHO sessions since January.

I'd like to introduce my colleague Commander Sabrina DeBose safety team lead in CDC's Division of Laboratory Systems, who will be facilitating today's session and the next couple sessions as well. Sabrina.

Sabrina DeBose: Yes, thank you for the introduction, Aufra, and thank you all for joining our sixth ECHO session. The topic for this interactive discussion is Laboratory Acquired Infections. Today's subject matter experts are Dr. Sheldon Campbell from Yale School of Medicine and Veterans Affairs Connecticut Health Care and Dr. David Peaper from Yale School of Medicine and Yale New Haven Hospital.

To foster a stronger sense of community and facilitate networking among the biosafety professionals, we encourage everyone to turn on their cameras during today's session. We understand that it may not be feasible for everyone, so please feel free to join with or without your camera. But our aim with the ECHO Biosafety Project is to connect names and faces and to build a community of practice.

I would like to ask everyone a quick icebreaker question. As we know, 4th of July is just around the corner, and we're all excited. What exciting plans does everyone have in store to celebrate the Independence Day this year? So please come off– come on the camera. Turn on your mics, or put it in the chat. But we want to hear about your plans and what everyone has to look forward to.

Let's see. I like it. I will wear stars and stripe socks. I like that, very fashionable.

Camping. Is it glamping? Or are you going to go officially and go camping? Barbecue and fireworks, that's always a good one. Alright, glamping. I like that, Courtney. That would be my speed as well. So would anyone like to come on camera and just say Hello and give us some information about your plans?

Alright, in the meantime, I'll keep reading. Oh, watch fireworks from my cousin's backyard and play lots of board games. Yeah, board games, always fun. Alright, well, what we will do is go ahead and keep going. But we want to thank you all for joining in our discussion today.

Let's see. Let me share my slide here. Technical difficulties. There we go. Alright, so thank you all for participating in our icebreaker.

Before we continue, I'd like to address some technical aspects of our ECHO Biosafety sessions. Please utilize the video capabilities from your device for this session. Currently, at this time, we have all audience microphones are muted, but we're asking, when engaging in discussions please unmute yourself to speak.

If you're experiencing any technical difficulties during the session, please send a private chat message to George Xiang, who is labeled as CDC ECHO Tech, and George will do his best to respond to your issue. If you're connecting to Zoom by phone only, at the time of discussion, please introduce yourself by stating your name and institution before speaking.

And you're wondering, how do these ECHO sessions differ from any other presentations? These presentations are different from webinars in that the main feature is the discussion of cases or clinical laboratory challenges. Our subject matter experts aim to share some applicable solutions that can be implemented in your individual laboratories. We encourage your active participation by sharing your knowledge and expertise. We know each laboratory is unique and your skill sets are unique, so your contributions to the discussions are truly valuable.

So here's a brief overview of today's discussion. I'll introduce our subject matter experts, which is Dr. Sheldon Campbell and Dr. David Peaper, who will provide a didactic presentation and a real case discussion. Then Aufra will summarize today's discussion. Closing comments and reminders will follow this, and we will adjourn the session.

Today's session is being recorded, so if you prefer not to be recorded, please disconnect now. We also have closed caption is provided for this session, and you can find the link in the chat box. After today's session, the transcript, the audio recordings, the presentation slides, and other resources will be posted on the DLS ECHO Biosafety website.

Now, it is my pleasure to introduce Dr. Sheldon Campbell and Dr. David Peaper. I'll read their bios.

Dr. Sheldon completed his MD and PhD degrees at Baylor College of Medicine in Houston and his residency and fellowship in laboratory medicine at the Yale School of Medicine. Dr. Campbell is Director of Microbiology, Chemistry, and Point-of-care Testing for VA Connecticut. Dr. Campbell's research interests include the education of pathology residents and medical students, point-of-care testing, and laboratory utilization.

Dr. Peaper completed his MD and PhD degrees at Yale School of Medicine. He remained at Yale for residency training in Clinical Pathology and a Medical Microbiology fellowship. Dr. Peaper directs the clinical microbiology laboratory at Yale New Haven Hospital in addition to the Virology Reference Laboratory at Veterans Affairs Connecticut. His primary research interests are studying the impact of clinical decision support tools and laboratory workflow modification on the delivery of microbiological testing and patient care. At this time, Dr. Campbell and Dr. Peaper, I'll turn the floor over to you.

Sheldon Campbell: Cool. I believe I'm starting because we always save the best for last. And let me turn this into a slide show. And is this the right view for you folks?

Sabrina DeBose: Yes.

David Peaper: You should swap them. Oh, there you go. That's good.

Sheldon Campbell: Swapped. OK, there we go.

David Peaper: That's good.

Sheldon Campbell: So Dave and I have divided this up into halves, and I've got the first half. I'm going to give an overview of laboratory-acquired infections and medical laboratory biosafety. Hopefully, at the end of this, you'll be able to discuss potential routes of acquisition of infections in the clinical laboratory.

Much of the literature on laboratory-based infections is actually in research labs. And we in the clinical laboratory are different. People keep telling us that. I'm not sure what they mean by it, but anyway. Look at some historical and known literature on this talk about the gaps because I think we've got as many gaps in our biosafety knowledge for the clinical lab as we have actual knowledge, particularly in the light of both the Ebola 2014-15 outbreak and COVID. And then I'll turn it over to David for a couple of highly illustrative cases.

Emerging infections and infections in general are not a new thing in human history. They've been around for as long as humanity has been around and keeping records. This is a picture of the seventh plague of Egypt. So infection in clinical laboratories– far more is unknown than is known. But it's very clear that clinical laboratory workers are at risk for infection.

It is little recognized that, in fact, during the HIV/AIDS era, laboratory workers were at a disproportionate risk for infection. Less than 3% of all healthcare workers are laboratory professionals, but we interact with many more patients than other healthcare workers do. And if you take all of the documented and possible occupationally acquired HIV infections between 1981 and 2010, about 30% of those involved clinical laboratory workers.

Well, we're only 3% of the healthcare workers. So on a per-capita basis, the risk to a clinical laboratory worker in that era was 10-fold higher than for other healthcare workers, including people like nurses. There are five times as many nurses as there are clinical laboratory workers. But we have made progress since then. And in the SARS outbreak, there were no– there was no SARS as far as any of us can figure out in clinical laboratory workers. So was that progress or luck? Still, I think, a little bit open to question.

But particularly in the microbiology world, laboratory-acquired infections have been a real thing. The Rickettsia prowazekii, the agent of epidemic typhus, killed both of the people that it was named after. Ricketts– Howard Taylor Ricketts was killed by the disease in Mexico. Stanislaus von Prowazek in Germany and many other researchers, particularly on infectious diseases, had been killed by their pathogen.

But how about in the clinical laboratory? The real answer to how many clinical laboratory-acquired infections are there is we have no idea. There is no reporting system. In fact, there are disincentives to report laboratory-acquired infections in a litigious and somewhat error-intolerant environment. Been a couple of attempts to figure out how much of this is going on.

In 2002 through '04– 20 years ago now, and things have changed in 20 years– Ellen Jo Baron and Mike Miller did a survey of clinical laboratory directors via the clinical microbiology listserv, ClinMicroNet. And a little over 80 laboratories answered. 33% of them reported at least one laboratory-acquired infection. 41 total bacterial laboratory-acquired infections were reported, many of them Shigellas, but also Brucella, Salmonella, S. aureus, Neisseria meningitidis, enteropathogenic E. coli, and C. diff.

This is a horribly inadequate way to survey this stuff. It was a completely voluntary survey highly biased toward large laboratories that have a full-time clinical microbiology director and with no notion of how complete the responses were. But it's clear that things are happening. And if you sit down at a session on laboratory safety at a meeting, people exchange a lot of anecdotes. But there is no surveillance and no systematic experience. It's clear that these infections do continue. Every so now and then in MMWR, there will be laboratory-acquired infections or even outbreaks of laboratory-acquired infections reported, predominantly again, however, from research laboratories.

We can at least assess where clinical laboratory-acquired infections might come from and start thinking about risks in the clinical laboratory by just looking at our operations. And the routes of exposure in the clinical laboratory can be inhalation. We do have procedures that produce aerosols. We centrifuge things. We mix, we sonicate, we vortex, we blend. And those are all things that we do intentionally. Then there are things that we may do unintentionally like spills and splashes as well. But we know that these procedures produce droplets and aerosols.

We of course are aware of and spend a lot of time thinking about the inoculation route via needle sticks or other percutaneous exposures. It's clear that ingestion is also a potential route of exposure, either through splashes or through practices that are problematic, or through the fact that microbes are small and there are a lot of them and we contaminate keyboards and telephones and we're not anywhere near as careful with those things as we maybe should be. And then finally, similar to inhalation, we can splash ourselves with larger droplets or gross volumes of liquid and contaminate skin and mucous membranes directly.

Risk presumably– well, risk is found in all the phases of clinical laboratory practice from the pre-analytic phase where sample collection, transport, reception and unpacking, centrifugation, uncapping, aliquoting, transport within the lab, and transport to reference labs all carry risk. On the analytic phase, all the different sections of the laboratory actually have their own risk profile driven by their instrumentation and their processes. And then even in the post-analytic phase, there are places where risk occurs– waste management and sample storage and retrieval.

In the analytic phase, the different sections of the laboratory do have meaningfully different levels of risk. In chemistry there are these complicated auto-analyzers. They've got multiple sampling stations. They're aliquoting events on the instrument that are not always obvious, especially to people who don't know the instrument intimately. And finally, there's a waste pathway where caps may or pipette tips may travel.

Most chemistry systems can't perform closed-tube sampling so that there's a risk associated with decapping. They of course, require frequent periodic maintenance and service, like all our big laboratory instruments do. And they're really expensive, so if you contaminate one of them you can impact care of large numbers of patients.

In the blood gas world, well, the samples come in a syringe. Do we love that? Not too much. Fortunately, it doesn't come in a syringe with a needle, mostly. But it's a very labile sample. And so there's no time to sit around and have elaborate precautions with it. We tend to work pretty rapidly with blood gases. In hematology, the situation is very similar to that in chemistry where you've got a big ole analyzer that's manipulating the blood in a lot of complicated ways. And in addition, there's often glass slides produced by a hematology analyzer.

In microbiology and bacteriology, there are all sorts of risks with the fact that we're amplifying organisms in culture, some of them with very high pathogenic potential like Shigella, which got a really low infectious dose. We don't always know what's in there. If we get a Brucella or a Francisella or a Yersinia pestis, nobody tells us that that's in there until we tell somebody else because we're the ones who figure out what it is.

Viruses in addition survive pretty well in culture media. There are old studies on blood culture bottles that show that HIV, for example, survives for days in blood culture bottles. So we can assume that less fragile viruses– and HIV is quite a fragile virus– probably also survive in culture media as well. And then finally, in microbiology we're now automating things, and we've got complicated analyzers, too.

In virology, many of the similar things are occurring as in other automated areas of the laboratory. Virology is becoming increasingly automated and increasingly molecular. Few virology laboratories now are doing viral culture, but if you're still doing viral culture, you could easily grow an emerging virus. In molecular diagnostics, in addition to the whole complex analyzer problem, we also have a lot of fairly manual methods still, especially in higher-complexity laboratories. And we've got all sorts of issues in many parts of the lab, but especially in molecular and virology, of how to validate and deal with EUA tests for dangerous rare pathogens that we may not have access to. How do you validate an Ebola test?

In transfusion medicine, there's still a lot of manual testing. There is no sealed rotor blood bank centrifuge, as far as I can tell. And the gel and instrument methods, similar to other automated laboratory processes, have very poorly characterized risk profiles.

So a whole swarm of folks got together and looked into some of these biosafety issues raised by Ebola published just as COVID was taking off, so it doesn't include much COVID experience. But the risks associated with biosafety issues really fall into four groups. And those are biosafety gaps that are common across all the clinical laboratories– gaps unique to specific areas, systemic gaps in biorisk management and processes as applied to clinical laboratories, and some specific lessons learned.

This is a big old paper. And I'll try and abstract the most important conclusions, but I do recommend it to you as a resource for thinking about problems in clinical laboratory biosafety. So gaps common across all our labs include the fact that we've got no direct control over how specimens are collected and transported.

You know, I'm not going to have you raise your hands, because I can't see you right now, but raise your hand if you've never gotten a leaking specimen through the tube system. I'll bet nobody is raising their hand unless they don't have a tube system. Our knowledge of instrument contamination during routine use or during use with highly pathogenic microbes is limited. And those risks may be underappreciated. We do know that certain parts of, for example, chemistry analyzers are contaminated by patient specimens during use.

We also really don't have a great grasp of decontamination of laboratory instruments. And that persists into the era of the post-Ebola and post-COVID era. There's actually discrepancies between the designation of Category A infectious substances and the very vast range of waste materials generated by clinical labs, some of which also have toxic chemicals in them. And so the whole area of waste management in clinical laboratories associated with biosafety is still a little bit ambiguous.

We are at the bottom of the list of people who get guidance and training in PP and often in availability. There are differences in the way you should be wearing PPE for direct patient care and PPE in a clinical laboratory environment. And that's something that we have not spent enough effort on. It's often hard for providers of laboratory safety training, such as CDC and the Division of Laboratory Services, to get evaluation data beyond learner satisfaction because we're not very good at assessing our biosafety practices. And data are lacking, really, on to what extent laboratories conduct monitoring and evaluation of their practices.

In specific areas, blood banks have the unique biosafety concerns I mentioned earlier and provide really critical supporting care. Core laboratories have got automation and instrumentation. And the risk profile of a core laboratory is just completely different from any research laboratory. And most of the biosafety literature comes from research laboratories.

Microbiology laboratories include all sorts of specific sort of platforms– automated systems that are very hard to change to manual. And we do manual system processes like make malaria smears and gram stains. And anatomic pathology involves big volumes of tissue, often with unknown pathogens, especially in the autopsy setting.

In addition, we really have gaps in our management processes. Clinical laboratories often don't know about what's in our specimens. Most of the research guidance is built around known hazards. We don't have known hazards. We have unknown hazards.

We often focus on rare events and preventive efforts. And it's unclear how you should evaluate those efforts. What is success in biosafety, not having an event or improving your processes for preventing one? And so there really is not a full management cycle in most laboratories– assessment, mitigation, performance, evaluation. There certainly is not in mine, even though I pretend to be an expert here.

There's no surveillance system nationwide or on any other level for laboratory-acquired infections. There's not much research and publication on clinical laboratory biosafety. The regulatory framework is weak and limited. There are just a handful of things in, for example, the CAP checklist, and they're mostly general on safety. It's nonexistent for waived labs, and waived labs are an increasing number and importance of labs in the United States. And biosafety guidelines really don't address risks in specialty areas of the laboratory at all.

We learned several things out of Ebola in 2014-15 that there were gaps everywhere– pre-analytical, analytical, and post-analytical– that we needed to carefully and thoughtfully limit testing without killing patients– and that's not a trivial thing to do. We have to develop testing menus for patients under investigation for serious viral diseases that emphasize the need to maximize diagnostic yield while minimizing risk.

There are ethical challenges. How do we balance the duty to provide laboratory services for routine patient care and personnel protection? How do we balance the duty to provide care for patients under investigation with the duty to provide care for the rest of the hospital population? And we still get conflicting guidance from different sources about safety issues that really confuses laboratory people.

I'd add a few more lessons from COVID-19. These haven't been vetted the way the peer-reviewed publication has. Data is still really important. There was one early paper that suggested there was little viruria or viremia in COVID. Thank goodness. It was not very systematically studied, however. And I was not real comfortable with that single publication early on in the pandemic.

I don't know what we would have done early in COVID if there had been a lot of virus in urine or serum. And there certainly could have been. SARS-CoV-1 had a lot of viruria. Fortunately, SARS-CoV-2 did not. And something really important out of COVID that didn't come out of the Ebola experience is that other workers are part of the risk. In a pandemic not of small numbers of severe disease but in large numbers of moderate or moderately severe disease, person-to-person transmission in the work environment may be more dominant than strictly laboratory-acquired infection.

So we know that these risks exist. What now? How do we address these needs? Well, we need more research. We need to research hazards associated with modern clinical laboratory equipment. It's unclear who would fund that. It's hard to imagine too many manufacturers funding research on reasons not to buy their equipment.

We really need some kind of surveillance system and clinical laboratory safety improvement programs. We need to assess the impact of delays in testing on carers, care for patients at risk of emerging infections, and how we balance the risk to patients and staff. We need better standards in risk assessment and preparedness and safety and instrument safety and decontamination.

And I think hopefully, the path forward includes more national oversight of clinical biosafety, improvements in instrument design, better training, guidance in managing laboratory workflow for cases of emerging infections at all the phases of testing, creation of a service– surveillance system for LAIs, and all the research and all the processes and things that we need to make more informed decisions about biosafety, especially in the context of emerging infections, and even guidance on waste management and specimen storage.

With all that general stuff, I think now it's time to turn this over to David, who will give some concrete and hopefully illustrative examples that will focus our discussions in a little more specific way.

David Peaper: Thank you, Sheldon. Let me get my screen up. You can see my screen?

Sheldon Campbell: Yep.

David Peaper: Great. So with that background provided by Dr. Campbell, I wanted to talk about a couple of cases, fortunately, of lab exposures without infection involving Brucella at Yale New Haven Hospital that occurred several years ago. But I think some of the information we've discussed around COVID is going to be slightly germane to the conversation as well.

So for the goals, I want to talk about these two cases, talk about some of the factors that we identified within our analysis of these incidents within the laboratory that we thought were– that was associated with these exposures. In our investigation and discussions, we identified some areas where we think we could provide some clarity and some of the guidance around when you should be thinking about these pathogens, when you should be entering into the "rule out or refer" pathway, which I'll discuss briefly. And then I have a little bit of information on the Select Agent Program and where that fits into this a little bit.

So the first case– and this was a major case we had a number of years ago– the presentation was an eight-year-old child who had multiple healthcare encounters over the course of a couple-week period, basically, from late November to early December. And the most pressing events of this case actually transpired over the long Thanksgiving weekend because of course they did. So we have an eight-year-old who's got multiple visits to his primary care provider, as well as an offsite emergency department, for fevers, vomiting, abdominal pain, and loss of appetite.

Some of the earlier laboratory evaluations demonstrated some slight abnormalities. There was not much of a substantial infectious disease workup at that time. He was clinically diagnosed with a viral syndrome. Then on the end of November on 11/21, he presented to our main emergency room at our main campus still having fevers, abdominal pain, near-daily vomiting, reporting now thigh pain as well as dorsal pain when walking. And his mother reported that the family traveled to Egypt in August, whereas in previous encounters they had denied any travel.

At the emergency department, he was found to have some anorexia without substantial weight loss. He was more lethargic, fatigued. He had some fevers and chills. He had some diaphoresis. He had a sore throat but with no congestion or rhinorrhea or sneezing. He did not have any cough or dyspnea. He did have abdominal pain, nausea, and vomiting, but there was no occult blood noted or no diarrhea reported. He did not have any GU symptoms. He did have myalgias, and he did have some headaches.

On exam at that time, he was afebrile. He had later documented fevers. His other vital signs were otherwise in the normal range for an eight-year-old, generally well developed and well nourished. Exam was relatively unremarkable except for some hepatosplenomegaly with some right upper quadrant tenderness. He was also found to be warm and maybe a little dehydrated but without jaundice.

So he was seen in the emergency room. They did some workup at that time. He was discharged from the emergency room, but he did have blood cultures collected at that time. He had some blood cultures collected again the next day.

He then re-presented in early December and had some imaging studies performed that showed some liver lesions. And then there was an MRI that was suggestive of osteomyelitis, and he continued to have GI symptoms. I'm obviously summarizing a lot of clinical history here in a couple of slides.

So at that point, the primary clinical differential did not include brucellosis but was primarily comprised of syndromes that could be thought to be found based upon travel to Egypt. And now that we are about 10 weeks out from that trip, these included things like malaria, typhoid, yellow fever, hepatitis A, Zika, dengue, and schistosomiasis. And so that's where the infectious disease workup was focusing at this time.

From the original blood cultures that were collected, the first set that was collected, two out of two bottles grew a Bacillus species that was not anthracis as well as coagulase-negative Staph[ylococcus]. The second set of culture bottles that were collected a day later, one bottle grew Ochrobactrum anthropi as identified by MALDI-TOF mass spectrometry at a 99.9% confidence identification. We'll come back to that.

The stool PCR performed at that time was positive for Yersinia enterocolitica. The other workup was unrevealing, including malaria testing, hepatitis testing, schistosoma testing, CMV, EBV, et cetera. At that point, the noninfectious workup was also unrevealing. And while the laboratory was not alerted to the concern over Brucella, there was an order for Brucella serologies that were ordered on the last day of hospitalization the day of discharge.

So using the retrospective scope, some of the thoughts around Brucella at that time were that he has this hepatosplenomegaly, he has fever, and he has these bone lesions that are consistent with osteomyelitis, potentially some degree of a disseminated illness. Factors that were thought to not necessarily exclude but to weigh against were that at that point, other than travel to Egypt, there had been no specific high-risk exposures elicited despite repeated asking about dietary exposures, environmental exposures, animal exposures, and the like.

It was also felt that the emergence of symptoms about 10 weeks after return from Egypt was very unlikely to be brucellosis, given that the time frame is typically considered in the two- to four-week range. From a microbiological perspective, the discussion was around the fact that the blood cultures that we did grow Ochrobactrum anthropi from grew within three days of inoculation or collection that when we reviewed the culture plates, that there was adequate growth in the cultures within that first read, within that first incubation, which was thought to be, once again, atypical for Brucella, and that then we had a very high confident identification for MALDI-TOF. And so this seven or eight years ago when MALDI was still a shiny new object that, in some respects, could do no wrong.

But unfortunately, the Brucella antibody serologies came back at a titer of 1 to 1,280. And at that point, the lab was notified that this was now much, much, much, much higher on the differential diagnosis. And so when we were notified, we still had some pending cultures, which we flagged all of those within the laboratory for special workup, special consideration.

The blood cultures from 12/5 were found to have grown small gram-negative rods that we could not rule out Brucella. Those were sent to the state lab, where they were positive for Brucella as well. Those were handled appropriately within a biosafety cabinet.

But this then begged the question of, well, what about that isolate of Ochrobactrum anthropi from 11/21? We retrieved that, and that was also positive for Brucella at that point. And species identification by CDC was consistent with Brucella melitensis. At that point, we undertook an exposure investigation.

Slight video clip of our reaction in the laboratory to that testing. There is sound involved if you have your speakers on. That was what I sounded like during that initial phone call. So we do have an audience response question.

In writing some of these questions, they're not written as a this is the definitive, best answer, but wanted to see what folks thought of some of these questions. And so in your laboratory or in your opinion, what criteria should a clinical lab use when we're assessing the risk of a culture containing Brucella?

Should we look at the presence of gram-negative rods with no growth on MacConkey? Should we look at gram-negative rods, no growth on MacConkey, plus some growth characteristics on blood and chocolate, particularly rate of growth, size of colonies, and the like? Should we look only at the gram stain reaction? Or is there the Potter Stewart definition of "you know it when you see it" when you encounter this in the laboratory? So we'll take a couple seconds to answer these questions.

George, I assume you'll pop up the response when you hit your threshold.

George Xiang: Yes.

David Peaper: That's great. Great. So I think it's a really nice mix here of that gram-negative rods with no growth on MacConkey with those growth characteristics on blood and chocolate, with some folks saying gram stain reaction and others saying just gram-negative rod with no growth on MacConkey.

So this is from the current APHL guidelines describing Brucella and when we should be worried about Brucella. I think there's a couple really interesting points here. So number one, we talk about the gram stain, faintly staining, nonclustered, go through some of the biochemical testing, which in order to do the biochemical testing, you have to have enter it into the rule out and refer pathway. And I think one of the major points of conversation that we had in our lab was, well, how do when to enter into that rule out or refer pathway?

And so we had been looking at slow growth, saying slow growth seen on blood and chocolate with pinpoint colonies at 24 hours, easily visible, discreet, white nonhemolytic colonies at 48 hours, poor to no growth– poor to no growth on Mac as well. And so this is saying 48 hours on blood. This is 72 hours on chocolate.

I can tell you from our case experience because I did look at the plates, this is probably what we saw at the 24- to 48-hour mark. There's more growth than this at the 24-hour mark. And so that was one of the factors that we looked at and said, well, this can't be Brucella. It just grew way too well for us.

Moving down the diagnostic algorithm and so once again, when you enter into this, it's you've got the gram stain morphology, faintly staining, nonclustered gram-negative coccobacilli and have a gram stain coming up. Then is it positive on blood and chocolate? Is it slow growing after 24 hours, and is it not growing on Mac? And is it slow growing in automated blood culture system?

And so once again, retrospectively going back and saying, well, we did see the blood on chocolate, we didn't see growth on Mac, question is, what exactly is a poorly growing colony after 24 hours of incubation, and what exactly is slowly growing in an automated blood culture system? Considering incubation for up to two to three weeks, these grew in three days. So this grew within completely our standard. In fact, both of the cultures that turned out positive for Brucella in this case.

So in our– following this, the question was, should we be entering the rule out or refer pathway or not? And retrospectively, I think we should have. But it's very hard, I think, to look at all of these and align them with our particular case.

Looking at the ASM guidelines for Brucella, once again we've got similar findings. Look at the gram stain, look at no growth on Mac. But I think one of the things in here that starts is talking about when you want a subculture. At this point, you're already supposed to be thinking Brucella because if you're going to subculture and spot it with a Staph[ylococcus] aureus for satellite testing, you're already thinking Brucella rather than entering the pathway sooner on. And so that's one of the points I'll discuss in a second.

So when we look at these findings from 11/21, 11/22, the order one had aerobic and anaerobic bottles growing the bacillus and the coag-negative Staph[ylococcus]. The order two was what ended up growing the gram-negative rod within 60 hours. And as I said, we had pretty much adequate or unremarkable plate growth on the blood and the chocolate the next day, enough to give us a MALDI ID of Ochrobactrum anthropi at 99%.

I just want to quickly go back and point out here that Ochrobactrum anthropi is actually not on the list of common misidentifications. Even still, it is listed in other places in the clin micro handbook. We did see some facultative anaerobic growth upon review of the work card. It was documented that there was no growth on Mac, but it was also not handled in those BSL-3 conditions.

So another quick response question here. So have the luxury of having a relatively new laboratory. It's about 10 or 11 years old now. We're up on the sixth floor. We have some great views. We're not relegated to the basement. But it is more of an open-concept laboratory. That is, there's not clearly defined room, clearly defined barriers among workspaces.

And so this was part of the conversation that we had was, how do you assess for the exposure risk for something like this, where there's a risk of aerosol transmission when you handle an open plate In a more open-concept-type laboratory? And so how would you approach this if you had to determine who's at risk for post-exposure prophylaxis and follow-up testing? Would you say I really have no idea, I'm going to consult with CDC or a local infection prevention and occ health? Would you say, well, if there's really no barriers then the whole lab's considered exposed, or would you come down with a hard marker of 5 feet, say?

Sabrina DeBose: We have a few participants, and we're waiting for just a few more before we close the poll.

David Peaper: Great. So we got a whole variety of responses. In our case, I would say we certainly discussed with infection prevention and health locally but ultimately conferred with the CDC. They ultimately recommended a 5-foot radius for evaluation for high-risk exposure for those needing post-exposure prophylaxis and follow-up testing. And then anyone within the laboratory that day could proceed, could pursue post-exposure prophylaxis and follow-up testing if they were higher risk or if they elected to.

I think what's interesting is we have now just come off of three years of talking about maintain your distance of 6 feet, what is an appropriate droplet radius, what's an appropriate aerosol radius, and the like. And ultimately, the guidance was on 6 feet versus 5 feet. And so I think it's interesting that at that time, we were advised on a 5-foot radius.

So based on that isolate from 11/21, there were a number of employees from Yale New Haven Hospital as well as Yale University that were present in the lab where the sample was not handled under BSL-3 conditions due to Brucella not being identified or suspected at that time.

We actually also had a couple of visiting medical students who were involved. So the recommendations at that time were that anyone in the lab where a Brucella worker who had already handled the specimen or were within 5 feet of the specimen were high risk and should have three weeks of prophylaxis with doxycycline and rifampin, and that anyone further away could discuss prophylaxis or consider prophylaxis if pregnant or immunocompromised.

Additionally, those who were exposed considered high risk all had sequential serum testing at 0, 6, 12, 18, and 24 weeks to see if there was evidence of seroconversion, as well as daily self-fever checks for 24 weeks. There were weekly symptom watch, required phone contacts from a healthcare provider for 24 weeks. At that, based on this incident, the total number of exposed workers was 21 total individuals. And there were no documented cases and no cases of laboratory-acquired infection from that particular incident.

Outlined here, there were some higher-risk individuals who had no adverse outcomes. There were some folks who did have to change their prophylaxis regimen due to intolerance to some of the medications. And we did under view our practice. We did not do a formal root cause analysis but did review our practices.

And so part of this is entering into the National Laboratory Response Network comprised of this pyramid with sentinel labs, such as hospital-based laboratories, at the base that are– where most all of the specimens were going to be coming in where a potential agent is selected. And the mantra of– the pathway is the we should rule out or refer. And so we should follow our standard rule out testing guidelines. And if we cannot rule out a select agent or a potential severe pathogen such as Brucella, then we need to refer up to the next level. In our case, that would be our Connecticut state lab. In other cases, there may be other labs that may act as the reference lab before referring up to the national laboratories. One of the key things, once again, is when the thought of entering into the rule out or refer pathway first enters the mind of the laboratory technologist.

So back gets into the topic of select agents. And so I think here on the right are the list of select agents. And those that are highlighted yellow are agents that could be reasonably recovered within a clinical microbiology laboratory. I've not highlighted all of the plant pathogens. But you see some of our heavy hitters here, such as B. cereus biovar anthracis, Francisella, Yersinia pestis, Bacillus anthracis, our Brucellas, our Burkholderia mallei, and [B.] pseudomallei, and the like, as well as botulinum neurotoxin-producing species of Clostridium.

The rest of these that aren't highlighted are not things that would not be typically expected to grow within a microbiology laboratory under routine testing. So these select agents are things– are toxins, viruses, bacteria, and fungi that have the potential to pose a severe threat to human and animal health, to plant health, or to animal and plant products. Often use this interchangeably with the phrase "potential agents of bioterror."

So what's interesting is I just highlighted most of the agents are not going to be necessarily routinely identified through laboratory testing without undertaking specific testing. But of those bacterial agents, pretty much all of those could be expected to grow within a clinical laboratory using standard, routine clinical laboratory protocols. And so these are the agents that are considered to pose a risk to laboratory workers and the clinical diseases that they cause, Brucella being right here. And then a number of these have been in the news lately in terms of other potential exposures.

When we look at the major bacterial pathogens and differentiating, say, Brucella from Burkholderia mallei to Francisella tularensis, one of the key factors that stands out is either no growth on MacConkey or slow growth on MacConkey, and all of these requiring some degree of either BSL-3 practices or BSL-2 practices with higher level of PPE. When we're reviewing slides, Sheldon asked what's the difference between poorly staining and faintly staining. I'm not 100% sure, but there is a differentiation there.

But looking at gram stains– so this is a gram stain, actually, of that second case that I'm going to present in a second where this is the Brucella in a blood culture from the gram stain. Here's our standard E. coli as well as Pseudomonas. And you can see very clearly the almost grains of sand or very faint staining that you can see here– very hard to see even under 1,000x magnification.

So in this particular case, once Brucella had been identified from the patient, they went back and had a further discussion. And at that point, the family said, well, when they were in Egypt, they may have been on a farm playing with goats and camels. He may have had some unpasteurized dairy throughout his stay in Egypt. The patient ultimately received six weeks of antibiotics, but he did have some adverse reaction to the antibiotics. And he's been– he was monitored afterwards for signs of recurrence in that case.

So when we looked back at of the factors associated with this– and so I think a lot of it, once again, comes down to the idea of when does the laboratory know to enter into the rule out or refer pathway? And as Sheldon mentioned, as a laboratory, we cannot always rely upon our clinical colleagues, as much as we have a good relationship with them, to alert us that there is a concern for Brucella or Francisella, for example, based upon risk exposures. And in this case, the clinical team did ask, and the patient did deny that.

There were a number of factors that each in and of themselves did not exclude Brucella exclusively. But when added together, I think a cognitive bias makes it seem that, well, this certainly can't be the case why. We'd go, well, we think of Brucella as typically two to four weeks, and he didn't have symptoms for 10 weeks. We think of these as very slow growing in blood cultures, but they grew within a few days. We think of these as very slow growing on our culture plates, but these grew within a couple of days.

We think of MALDI not lying to us, but it gave us a very good identification for MALDI. And then further compounding issues around MALDI which, once again, this was relatively newish into the MALDI era, select agents are not currently in the MALDI database. I think there's an under-appreciation of that. So since all bacterial identifications are made in reference to a database, if the organism is not in the database, it's never actually going to give you that ID. We're talking about relatively unusual outcomes or relatively unusual pathogens, as Brucella.

But also, Ochrobactrum is not one of those common pathogens. So there is not necessarily familiarity among the laboratory staff to say, hey, this doesn't really fit like an Ochrobactrum in my experience, because most of our staff has not as much experience with Ochrobactrum as they would, say, with Pseudomonas or E. coli. In this case, we got the match to Ochrobactrum using MALDI.

I think within the laboratory, thinking about some of these human factors, our in-lab protocol lacks specificity when to suspect Brucella. So when looking back at things, the protocol might say, if the technologist suspects Brucella, do these other things, without providing very clear guidance around when to suspect Brucella upfront. And then I think based upon the polling response as well as, I think, what our experience was, I still think that there's some lack of clarity around that.

I think there's also a very strong cognitive bias against rare things, right? So the idea that, well, it's almost certainly not this thing does not equal it's definitely not. Stated differently, rare things are rare but not impossible. When we were looking back and having conversations around this particular case, there was some degree of circular logic being employed. Well, this can't be Brucella because if it had been Brucella, we would have ruled out Brucella, that's why it's not Brucella. And there was some incomplete communication around some of those areas.

This was compounded by the fact that it was a long holiday weekend. We had staff– multiple staff handling the culture across multiple shifts and days without fully adequate communication, and then an overreliance on verbal communication without necessarily going into the laboratory record and re-verifying the primary data was there growth on MacConkey or not instead of relying upon someone saying there is or is not growth in MacConkey.

I already mentioned the issues surrounding Ochrobactrum. And then once again, we had become– we are still very reliant upon MALDI-TOF for identification. We originally thought it was more likely to say I don't know rather than outright lie to us like some of the biochemical tests. But I think this case and others, as they have accumulated, does demonstrate that the MALDI does lie to us sometimes, unfortunately.

When we look at brucellosis in the U.S.A., we have staff in the lab who've been there for a very long time and could not remember the last time we had a Brucella case at Yale New Haven– probably at least 20 years or more. Typically pretty rare in the U.S.– around a hundred cases per year per report, with most of those being in California, Texas, Arizona, Florida, Georgia, Illinois. And so most states don't actually see a case in a year. So it's much less common, even, for a given institution or given provider to see a case in a particular year. There were certainly plenty of techs in the lab who'd probably never seen a Brucella case when this came through our doors.

And so just another quick audience response question– what would you do in your lab to avoid cases like this in the future? Would you re-educate the staff? You can answer as many as you want here– re-educate your staff, modify your protocols to explicitly list when an isolate should be ruled out, put a sticker system in place to communicate unusual gram stains across shifts and across staff, or any others? And if you had to confront this, I'd appreciate you entering some responses potentially into the chat box.

Sabrina DeBose: Alright. The responses are still coming in. And we'll give it a few more seconds.

David Peaper: Great. I think these are all things that we did undertake in our laboratory in immediate response to this. We did provide a lot more explicit information. We went with a substantially broader "if you see a gram-negative rod with no growth on MacConkey, enter into the rule out or refer pathway." One of the factors that was identified was there's a tech who reads the blood culture gram stain typically 18 to 24 hours before someone handles the plate, and there's not clear communication with the plate reader that maybe there was something unusual about the gram stain. So we started putting stickers on those things with unusual findings. And then we did substantially re-educate the staff on all of these issues.

But despite that, we did have another case. And I see someone in the chat has said potentially flag anything that takes more than 24 hours to grow since that's less common, look at sterile fluids body– blood, sterile fluids, and CSF under biosafety cabinet. And then the issue of blind samples will come up in a couple of seconds.

So this is our bonus case, which is not really much of a bonus. Almost exactly one year later, we had a 48-year-old who had just come back from Egypt to visit friends and family. He was there from to 10/1 to 10/13. He felt well while he was in Egypt. He had no sick contacts. He did not report eating undercooked meat or other food that was concerning to him.

He did stay in Cairo without any animal exposures. He felt well until he returned home and was his usual state of health until 11/1 or 11/2 when he developed headaches, fatigue, and malaise. He had multiple healthcare encounters over the next week with fevers, night sweats, and without diagnosis or resolution, presented to our emergency department, where we had blood cultures collected at 5:34 PM on the 12th.

They were positive at 12:18 AM on the 15th. And at that point, the gram stain read on those cultures were no organism seen, and our blind subculture procedure at that time called for subbing those to blood chocolate and an anaerobic blood agar plate. They were incubated and had– notice something's missing from that list. And so these were thought to have adequate growth the next day, and they were reviewed by the blood bench.

At some point, I was called over to look at a gram stain. All the plates were closed at that point thankfully. And it was PTSD of gram stain images seared into my brain. And so at that point, we had to stop the line, stop working up, enter into the rule out or referral pathway, sending it to the state. And that was confirmed by the state as Brucella the next day. At that point, we had approximately five laboratory workers who were considered exposed to these plates based upon the 5-foot radius that we had discussed previously.

So as you noticed, one of the major things that was missing from our blind subculture procedure was the absence of a MacConkey agar at that point. History was revisited. The patient continued to deny any contact with sheep, cows, unpasteurized dairy, et cetera. But one of his friends was available who spoke to him in Arabic, and they recounted their trip. And it was forthcoming that he did have some unpasteurized dairy in a monastery, I believe, when he was there.

So when we were reviewing this, we did find that because the lack of MacConkey agar in our blind subculture plate was thought to be a significant contributor to this, and so we've added a MacConkey plate to all of our blind subcultures within the lab. At that point, we also significantly rearranged the benches within the laboratory so that we exiled our blood culture bench to the far reaches of the laboratory so that there's actually– pretty much only the blood culture tech will be within 5 feet of any of the positive blood cultures at that time.

We also did put a biosafety cabinet directly adjacent to the blood culture bench to remove any potential barriers to moving that work to a biosafety cabinet for further workup in a rule out or refer pathway at that time. And since that time, we've not had any additional confirmed Brucellas. We've have sent off a substantial number of isolates for referral testing to the State of Connecticut. They may be frustrated with me, but it's following the rule out or refer guidelines as much as we can at that point.

So what are some of the challenges we've helped identify? So number one, select agents are rare. We're often not told that they're in the clinical differential. They're often not in the clinical differential. They may not be in the microbial differential either. Rare things are not impossible, but they are rare.

We know that these select agents should not enter the regular laboratory workflow. We should do the rule out or refer pathway. But often, these organisms are not considered before an unusual identification is made. We know that all IDs are made in comparison to a database, and so that if an organism is not in a database, it's never going to have been made as an identification. And most databases don't contain select agents.

You should not be testing these on commercial test systems. It's very clear. But if the procedure does break down, the potential for misidentification can further, I think, prolong and delay the identification of a potential select agent. We would ask our clinical colleagues, if our patients have comparable– or compatible travel and an unexplained fever and they're sending cultures to the laboratory, that the lab should be notified.

For the most part, we do get notified. We know when we are notified. We don't know when we're not. And I think that's always one of the challenges. There's always a conversation around, well, is there a clinical decision support tool or some EMR tool that we could use to help alert us to this? I think our experience with Ebola has made that a challenge.

By the time you are thinking of asking the questions with the degree of detail necessary to trigger an alert, you probably have already started down that differential diagnosis anyway. So there does not seem to be. But I think these are conversations. These are areas that we could potentially investigate in the future. So with that—

Sheldon Campbell: By the time you get that deep in the tree, you've already spent a couple of days working on more common things.

David Peaper: Exactly. So I'd like to thank everyone for their time and their responsiveness. Appreciate the comments in the chat as well as the responses to the polling questions.

Sabrina DeBose: Yeah. So we want to thank you all very much. We have some discussion in the chat. We can– I did see something from Marcia– no source of history provided to the microlab, and also provide blind samples or rule out organisms at random. So a couple of discussions in the chat.

But we do want to thank you both for your excellent and informative presentations. What we would like to do right now is open the floor for any ideas, questions, or additional discussion points from the audience. And we appreciate you all and encourage you to share your experiences and challenges on the topics that we just covered. So please remember to turn on your cameras and microphones and introduce yourself if there are any additional topics you would like to discuss or questions or ideas you all would like to share.

David Peaper: I would just echo what Marcia– on Marcia's comment about providing blind samples to rule out organisms at random. I think many of us are College of American Pathologists-accredited laboratories, and we may subscribe to the LPX survey, which is a great survey. But you know when the LPX survey is coming, and you know when you get it. And so I think the idea of doing at-random samples within the laboratory is something that we have discussed, and it fell by the wayside.

But I think that's definitely something that we want to re-initiate or initiate within our laboratory to help the techs just keep it on top of their mind. We've had a lot of turnover for staff. A lot of those folks who were exposed or involved in these exposures have left the laboratory and retired. And so I think that's a great suggestion. And how to actually execute that is, I think, going to be a little challenging but definitely something that's good.

Sabrina DeBose: Erin, we see your hand. Please chime in.

Erin Bowles: I really thank both of you for sharing all of the education today. It's a really daunting task for those of us who are trainers and have to try and reach out to the sentinel laboratories, to try and prepare them to handle these and prevent exposures. And despite all the efforts, it just seems like they still get caught.

So I will take some of these stories and remind them that the bugs, they don't always behave the way we expect them to. The fact that they grew much better in 24 to 48 hours, you would have expected that's not an invitation to put them on MALDI-TOF, which is pretty much everybody's ID system that they're using these days. And so they do get caught with these things.

Or even when they're doing rapid testing on blood cultures to try and get identifications and not working with samples inside biological safety cabinets, that's another area that they can have exposure. So I think the more that people share these– I really wish there was a better way to report these incidents so that people really realize that although it's rare, it's definitely not impossible, as you said in the slide. They occur– we have a couple every year in just Wisconsin alone. So I know other states are having the same thing. So thank you, though.

Sabrina DeBose: Thank you, Erin. We appreciate that feedback. I do have a question for you all, Dr. Sheldon and Dr. Peaper. How do you envision a national LAI reporting system operating?

Sheldon Campbell: Yeah. I mean, I'm not sure. I'm not sure how you'd build that kind of system. It almost has to be built into the laboratory regulatory framework, which CLIA doesn't mandate anything like that. I don't have a great feel for that. Note that if there were such a system, David's incidents wouldn't be in it, because there were no infections.

So do you measure near misses? It's a bigger conversation.

Sabrina DeBose: Right. Yeah, I agree. And it does take a lot of detail. I think this was a great audience to present that to saying we wanted to keep that community of practice and get people to thinking. So that's something we want to continue to keep on individuals' forefront. And maybe we'll eventually get there one day.

David Peaper: I think it would be interesting. I think it would be interesting to look at all of the reported– because these infections are reportable. But looking at– of the Brucellas that are reported annually, how many of those were associated with laboratory exposures? I think that'd be really interesting to help provide a little bit of scope around it– not necessarily lab infections, but how many lab exposures were associated with each case of brucellosis in the United States or each case of Francisella in the United States?

Sheldon Campbell: Some kind of survey sent to any lab that identifies one. Yeah, that'd be a good start.

Sabrina DeBose: We do have two discussions in the chat. One– let's see– is biosafety risk assessment a topic that is enforced by regulatory agencies?

Sheldon Campbell: The only regulatory agency that have a lot of knowledge of it is the College of American Pathologists. And there is– it's implied in your– you have to have a plan for handling emerging infections and high-consequence pathogens in your laboratory. And a risk assessment is implied in that, but I'm not sure that it's actually spelled out in the regulatory item.

David, do you know?

David Peaper: No. So I was just going to say, with respect to New York State, too, I think it's– they're not very prescriptive. It's not a matter of you must do these things so that you don't have lab exposures to Brucella. It's more of a general, you need to have an exposure plan, a biosafety plan, without necessarily providing as much detail as, I think, the lab, the microbiology laboratory needs to operationalize that type of plan. So it's often left to the lab.

And so ours might say if a microbiology technologist suspects Brucella, do the rule out or refer pathway within a biosafety cabinet using these procedures, without necessarily explicitly saying when someone should expect or suspect Brucella. And so I think saying, if you suspect Brucella do this, would meet the letter of the requirement without actually being super helpful.

Sheldon Campbell: Yeah. Again, the guidance in clinical laboratory regulation about safety is really general. And I think a lot of our experience is that it would be helpful if it were more concrete and specific.

Sabrina DeBose: There's also another comment in the chat from Marcia. It has been my experience that it is difficult for laboratories admin and managers to enforce risk assessments, as this is not in the inspection checklist.

Sheldon Campbell: Yep. Hard to get resources for anything that's not required these days.

Sabrina DeBose: Well, once again, we want to thank you both. This was an excellent discussion. And we thank you and thank everyone for participating. What would like to do now is invite my colleague Aufra to provide a summary of the discussion from today's session. Aufra, I'll turn it over to you.

Aufra Araujo: Thank you, Sabrina. And thank you so much, all participants and our speakers, for a rich discussion and awesome presentation today. I have some bullet points here about the discussion.

And one is need as it relates to risks in the laboratory. And basically, there are two needs, as Dr. Campbell mentioned, research and standards– research on hazards associated with modern clinical laboratory equipment, surveillance and reporting of LAIs, clinical laboratory safety improvement, and the impact of delays in testing on care for patients at risk of emerging infections, and also balancing risks to patients and staff processing the samples. The needs on standards– and we were just talking about it– the standards related to laboratory safety practices, how that's so much needed, one of them being clinical laboratory risk assessments, another one emerging infections preparedness and safety, and the other one instrumentation safety and decontamination.

Where do we go from here? And some suggestions of moving forward may include enhancing national oversight of clinical biosafety; improving improvements in instrument design– laboratory instrument design; more training materials and resources available for the clinical laboratory and public health labs, as well everybody doing diagnostic testing; also guidance in laboratory workflow in case of emerging infections in all pre-analytical, analytical, and post-analytical phase as well; creation of a surveillance system for laboratory-acquired infections.

That's a big challenge, but it's also a solution for monitoring and responding to LAIs, and so on.

Regarding some of the Brucella case that Dave presented– Dr. Peaper presented, it was easy to observe how complex is the algorithm for Brucella diagnosis– diagnostics. Also, in any open-concept laboratory, assessing exposure risk for post-exposure prophylaxis and follow-up testing, as well as consulting with CDC, infection prevention and control, and also occupational health are relevant.

The last point I'll make in the summary is that to avoid Brucella cases in the future, some points might be helpful. One, first, is to re-educate staff on characteristics of selected agents. Another is modify protocols to explicitly list when an isolate should be ruled out. And that could be based on the algorithm itself. And another suggestion in chat was to provide blind samples or rule out organisms at random.

I'll stop here, but I'd also like to invite the speakers in case they have something else to add. Thank you.

Sabrina DeBose: Thank you, Aufra.

Sheldon Campbell: Yeah, that was a very complete summary. I don't have anything to add.

David Peaper: Yeah, I don't have anything to add there.

Sabrina DeBose: Alright. Well, once again, we do want to say thank you to Dr. Peaper and Dr. Campbell. I would also like to thank all of our participants for taking part in this discussion today. We hope you found it valuable in the important work you engage in with your laboratories. We look forward to your participation in future sessions as we dive into specific laboratory biosafety topics.

Soon, you will receive an email containing a post-session survey link from the DLS Evaluation mailbox. It will take two minutes to complete. Please continue to complete these surveys, as your feedback is valuable to us. We read every comment and use your input in planning future sessions. We also share your responses with the presenters for their consideration. We appreciate your time and attention in completing the post-session surveys.

If you have additional comments, please send an email to dlsbiosafety@cdc.gov. I will share my screen here. Alright, so we are excited to have our next session in July. It will be on Tuesday, July 18, at 12:00 PM. The topic will be "Safety Implications of Diagnostics in Remote Areas," which will be presented by John Laurance from Alaska State Public Health Laboratories. Please visit the DLS ECHO Biosafety website to view all upcoming sessions. And if you have any questions, remember you can always reach out to us at dlsbiosafety@cdc.gov.

Now we will adjourn. And once again, we want to say thank you, and have a great day.

Additional resources and related publications

• Centers for Disease Control and Prevention. (2011, May 23). Surveillance of Occupationally Acquired HIV/AIDS in Healthcare Personnel, as of December 2010.
• Baron EJ, Miller JM. Bacterial and fungal infections among diagnostic laboratory workers: evaluating the risks. Diagnostic Microbiology & Infectious Disease, 2008 March; 60(3): 241-246.
• Cornish NE, Anderson NL, Arambula DG, Arduino MJ, Bryan A, Burton NC, Chen B, Dickson BA, Giri JG, Griffith NK, Pentella MA, Salerno RM, Sandhu P, Snyder JW, Tormey CA, Wagar EA, Weirich EG, Campbell S. Clinical Laboratory Biosafety Gaps: Lessons Learned from Past Outbreaks Reveal a Path to a Safer Future. Clinical Microbiology Reviews, 2021 June, 34(3): e0012618.