Hello, everyone. I'm Dr. Bhavita Gaglani, and I'm a practicing intensivist and an infectious disease faculty at Atrium HealthWake Forest Baptist University Hospital in Winston-Salem, North Carolina. And today, we'll discuss about the rapid diagnostics in pathogen identification in the ICU. Her objectives will be to describe current problems in rapid diagnosis and management of early sepsis in the ICU. We will review different diagnostic methods, which will include practical applications of next-generation sequencing testing for rapid pathogen identification. I have no disclosures relevant to this topic, and off-level medication and devices used may be discussed throughout the talk only for education purposes. So what are the problems we face in absence of early rapid pathogen identification? The first and foremost we worry about is delayed active therapy, which can hence lead to increased mortality. With a mortality risk of 40% to 70%, septic shock is the most common cause of death in the modern ICU. Delayed and inappropriate administration of optimized antibiotics is associated with progression from sepsis to septic shock and increased rates of adverse events, mortality, and health care costs. In one of the largest retrospective cohort study of the era as cited here, the authors studied more than 5,700 ICU patients with septic shock to determine the appropriateness of initial antimicrobial therapy based on the clinical infection site and relevant pathogens. And they found that inappropriate initial antimicrobial therapy for septic shock occurs in about 20% of the patients and is associated with a five-fold reduction in survival. The next problem we face is increased antimicrobial use, which can lead to increasing drug resistance. The CDC estimates that more than 2.8 million antibiotic-resistant infections occur annually in the United States, accounting for more than 35,000 deaths annually. Carbapenems in particular often represent the last resort for treating multidrug-resistant gram-negative bacterial infection. An early use of a carbapenem in febrile neutropenia is recommended for high-risk patients, which can have either an absolute neutrophil count less than 100 for more than seven days and or significant comorbidities or hemodynamic unstable patients in case of previous infection or colonization with ESBL-producing bacteria or in hospitals with high ESBL rates. As a consequence, the use of carbapenems in neutropenic patients has greatly increased, likely contributing to the rise of the carbapenem-resistant infections. In this systemic review, the authors looked into the currently available data on distribution, the characteristics, and the outcomes associated with carbapenem-resistant bloodstream infections in adult neutropenic patients, and they found that the highest risk of carbapenem resistance was associated with prior carbapenem exposure, as demonstrated in this forest plot. The other problems that we encounter is in confirming the type of infection itself. Best practice guidelines and quality metrics recommend immediate antibiotic treatment for all patients with suspected sepsis. However, little is known about how many patients given IV antibiotics in the emergency department are ultimately confirmed to have bacterial infection. In the single-center retrospective study of 300 adult patients with suspected sepsis who received broad-spectrum antibiotics in the ED, they found that 35% of the patients did not have a bacterial infection, and this can further lead to increased antimicrobial use and then delayed active therapy, causing the problems and outcomes as we had already discussed above. And last but not least, it delays intrinsic to culture-based assays. Now, a fun fact, Fanny Hesse, as pictured on the right, is best known for her work in microbiology alongside her husband, Walter Heast. Following her initial suggestion of using agar as an alternate to gelatin, they were instrumental in pioneering agar's usage as a common gelling agent for producing media capable of culturing microorganisms at high temperature. So the cultures go way back. Traditional culture methods to identify microorganisms are time-consuming and unfortunately are not as helpful in the early identification of sepsis. In order to find an optimal solution, the diagnostic method should be rapid and applicable to detect pathogens alongside drug resistance, preferably within three to five hours of patient admission. It should be capable enough to diagnose polymicrobial infections along with unknown and emerging pathogens. Furthermore, the results of the diagnosis should be able to provide appropriate decisions and antibiotic stewardship within a key time window of hours in order to limit morbidity and death. The procedures that are sensitive, specific, and quick for identifying the pathogen are therefore the major operational instruments for critical care units. In the next coming slides, we will briefly discuss about some of the diagnostic tools as classified in the scheme. The conventional methods use the selected cultures and medium in traditional approaches and detect basic aspects of bacterial identification. Molecular detection uses molecular base, which can be a DNA and or RNA, approaches to identify resistant genes as well as mutations and expression of these genes either at a targeted level or their whole genomic signature. Now, the gold standard for identification of bacterial and fungal pathogens have been culture. Despite simplicity and low cost, the turnaround time for culture-based methods can extend up to several days or even weeks, leading to delayed diagnosis, inappropriate antimicrobial use, and, in some cases, excess disease transmission in the hospital due to missed infections. Viral pathogens and some bacterial pathogens, such as mycoplasma pneumonia or lesion analymophilia, may be difficult to detect with traditional culture-based methods. Because empiric antibiotic treatment is typically administered as early as possible in patients presenting with infection-related symptoms, the use of culture-based identification might also lead to false negative results, as antibiotics can sterilize microbial cultures. Now, while standard blood and respiratory cultures are relatively inexpensive compared to many medical diagnostic tests in some countries, such as USA, the cost of labor and routine use of mass spectrometry for taxonomic identification have led to poor patient costs of several hundred US dollars. Now, when we look into blood cultures, in particular, which are generally used to diagnose sepsis because of the quantity of microbes occurring in the blood during infections, it requires continuous monitoring, especially with the conventional automated blood culture systems. So, starting from patient's blood draw till obtaining ID and susceptibility, and after which an optimized therapy can be determined, requires at least 48 to 72 hours. And this is where there are opportunities for rapid diagnostics. Thanks to the rapid blood culture identification systems, or we call the BCID panels, which are multiplex nucleic acid amplification assays that detects gram-positive and gram-negative organisms at the genus or species levels, and include resistant genes. There are different BCID panels that are available. It varies in amount of the bacterias that it detects and amount of resistant genes that it can detect. Important thing to understand about the BCID panels is that they lack an amplification step, and hence they need to be a threshold amount of bacteria present in the bloodstream for it to detect. These assays can detect common beta-lactam resistant markers having a very high positive and negative predictive value with staphylococcus species. It also has a good positive predictive value for ESBLs. It's mainly detecting the CTXM gene. It also detects good rate of CRE genes, including deoxa and DM. However, the negative predictive value remains more limited, and that's when the culture data definitely helps. It is also useful in detection of non-enzymatic resistant in pseudomonas with different pseudomonas species. With this genetic study available, what missing remains is the phenotypic accessibility profile in order to have rapid optimized therapy. A great observational study was performed at University of Maryland Medical System, which compared the theoretical antimicrobial treatment decisions guided by this two commercially available rapid diagnostic testing profile. And they evaluate the potential clinical utility using a framework that compares the potential antimicrobial therapy decisions by final organism identification and phenotypic susceptibility profile. I'm not going to go in too much detail for the sake of time. However, the studies identified E-PLEX-B CID to be able to identify more organisms than the variegin and hence having a good phenotypic susceptibility profile as well. This brings us to our modern molecular methods, and the first on the list will be MALDI-TOF. MALDI-TOF had been really a true blessing since it's used in past few years. It is a desorption ionization high throughput method that can accurately identify a large range of pathogens, including bacteria, yeast, filamentous fungi, and mycobacteria in minutes. This methodology is FDA approved for use on isolates grown in routine cultures, and protocols for testing specimens directly are commercially and widely available. Briefly, what it does is the samples are incorporated into a matrix and bombarded with a laser, which results in vaporization of the portion of the sample. The mass to charge ratio of the resulting molecular peptide fragments are then analyzed to produce a molecular signature, or even called a peptide fingerprint, for any unknown organism. This fingerprint is unique to the individual microorganism with peaks specific to genera, species, and strains. The test isolate signatures are then compared through a database of reference spectra to determine the identifications at species levels. However, the spectra can only identify species that are present in the database effectively, resulting into what we call blind spots. So there is a clear limitation of the method as misidentified reference strains have been reported to cause downstream identification problems. There are certain species that MALDI-TOF are still not able to differentiate in terms of the bacterial structure, for example, E. coli and Shigella, and their reference database spectra are proprietary to each manufacturer and have recently expanded. Another work in progress or under active investigation is its utility for antimicrobial susceptibility testing. Now, one of the first successful applications of MALDI-TOF mass spectrometry to detection of antibiotic resistance actually resulted from an observation that the hydrolysis of beta-lactam ring after exposure to beta-lactam antibiotics to beta-lactam producing bacteria could be revealed in mass spectra by the decrease of the peak corresponding to the antibiotic and appearance of those peaks representing its hydrolysis products. So far, only two commercially available kits with software for automated interpretation of the spectra have been authorized in Europe, which can detect carbapenemous activity or resistant towards third-generation cephalosporins in clinical microbiology labs. The next available methods on molecular levels are the PCR-based pathogen detection methods. It can be conventional, single-plex, or multiplex, and this allows target pathogen DNA sequence to be amplified, which is amplifying microorganisms' specific nucleic acid sequences, thus allowing the targeted detection of a set of predefined microorganisms, often within just a few hours. Despite the availability of many FDA-approved microbial tests, only a handful of PCR-based assays are clinically accepted and available in routine practice, and the less common organisms or the novel emerging pathogens or pathogen variants may still remain undetectable using such approaches. Some other limitations common to PCR diagnostics is that the interpretation of positive results with a negative culture remains challenging, especially in terms of differentiating true infection from normal microbiota, transient colonizer, or sample contaminant. On the other hand, without exhaustive multiplex detection of pathogens, a negative result does not rule out infection. Together with the lack of reliable antimicrobial susceptibility testing information, the PCR-based diagnostics remains constrained as an adjunctive test to the conventional cultures. Take an example, looking into respiratory tract infections, there are multiple commercial assays that are available with the pneumonia panel. Now, this was implemented at May 2020 in the midst of COVID-19 pandemic. At that time, we did not have the SARS-CoV-2 as target, although now the respiratory pathogen panel does include it. The pneumonia panel has a qualitative and semi-quantitative detection of resistant genes with multiple pathogens that can be found typically causing pneumonias. However, these resistant organisms are not always identified by cultures of the specimens in which the pneumonia panel detected the presence of the resistance gene, which suggests that the resistance gene may have been present in a non-pathogenic or low abundance organisms. There are some challenges that are faced is increased likelihood of identifying colonizing organisms, detection of multiple organisms that may or may not be identified in the culture, and the semi-quantitative results that are higher than what should be detected on the culture, and thus becomes a little difficult to interpret. Despite these difficulties, there are multiple studies looking into the potential benefit of pneumonia panel and have been identified in facilitating rapid identification of organisms and resistant genes providing results in approximately 75 minutes. It may lead to earlier initiation of appropriate therapy, prompt discontinuation of broad-spectrum empiric therapy, or rapid de-escalation to targeted therapy. This multicenter UK study, both platforms with the UNIVERO and the FLIM-LA of pneumonia panel performed similarly and were considerably more sensitive than routine microbiology, detecting pathogens in patient samples reported as culture negative. The increased sensitivity of detection released by the PCR does offer potential for improved antimicrobial prescribing. On the other hand, standard methods frequently fail to identify the lower respiratory tract infectious etiology due to the polymicrobial nature of respiratory specimens and the necessity of ordering specific tests to identify viral agents. The potential severity of these infections combined with a failure to clearly identify the causative pathogen results in administration of empiric antibiotic agents based on clinical presentation and other risk factors. In this study cited here, the authors examined the impact of the multiplexed semi-quantitative biofiber film area pneumonia panel on labs reporting for 259 adult inpatients by submitting their bronchiobular levate specimens in eight different US clinical centers between October 2016 to July 2017. In approximately half of all culture negative pneumonia panel detections, the patient had received antibiotic therapy in the 72-hour preceding the specimen collection, which could contribute to failure to recover this bacteria in the culture. An additional 43 percent of this specimen with culture negative deductions reported the presence of normal oral flora in the culture with a positive pneumonia panel. Further in the study, it was identified that for many patients, there was an opportunity for multiple antimicrobial modifications, including simultaneous escalation and de-escalation in almost 48 percent of the patients due to empirical utilization of multiple antibiotics in a single patient. This resulted in an average saving of 6.2 antibiotics day per patient. Looking into different samples, there are multiplex PCR in whole blood available as well. The whole blood multiplex PCR assays can detect more than 90 microorganisms at genus level, more than 30 at species level, and at least three drug-resistant genes. This provides results within six hours and so are very rapid to diagnose. However, there are studies looking into its sensitivity and especially their sensitivity compared to the standard blood cultures, and they found although they are less sensitive, they are highly specific. This assay definitely needs to be optimized mainly to improve the sensitivity and also include other significant microorganisms within the panel. Greater automation is necessary to facilitate introduction of this assay into routine lab workflow and to reduce further turnaround time. The further microbial detection limits of the molecular assay can also be improved. This assays are more likely to be positive than blood cultures in patients previously on antibiotics. Last but not the least comes metagenomics. This is an evolving branch in the rapid pathogen identification. It refers to a broad area of methods that rely on the sequencing of nucleic acids within a clinical sample in an attempt to identify pathogens of interest. This test take a pathogen agnostic approach and sequence all the nucleic acid present in a specimen with the hope of detecting the causative organism amongst any background contamination. This method differs from multiplex PCR methods in the utilization of a universal approach to the sequence with a wide variety of nucleic acid in the specimen, rather than an amplification-based approach which uses a limited set of pathogen-specific primers. Let's very briefly go through the simplified overview of metagenomic workflow, which is broken down into two main steps. The general metagenomic workflow begins with nucleic acid as extraction and it could be DNA and or RNA from the biological sample of interest. This step is then followed by library preparation during which the nucleic acids are fragmented and the short adapter sequences are further ligated into the ends of the fragments to permit PCR amplification and then binding to the sequencer flow cell. These samples are typically bar-coded to enable multiplexing and the whole process is called the wet lab. Once this is done and once the sequences are read, the data are then fed into bioinformatics pipeline for quality control, host subtraction, and taxonomic alignments, followed by identification and quantification of microbial species, and then functional analysis of the sequencing platforms, which is something happening in the dry lab. These platforms can be clinically used with a turnaround time ranging anywhere from six hours to several days, depending on the instrumentation, the degree of sample multiplexing, and its infrastructure. While there are different studies looking into the utility of next-generation sequencing, this observational study cited here had looked into 50 patients with septic shock, primarily with abdominal sores, and used 20 control patients undergoing elective surgery. They found that 33 percent of these patients with septic shock had positive blood culture at baseline, and 72 percent had positive NGS testing, identifying some of these bacterias that could have led to septic shock. Overall, 53 percent of patients would have had led to changes in more adequate treatment. There is a really nice multicenter study that's forthcoming, which is a prospective observational study for the first time investigating the performance as well as the clinical value of NGS-based approach for the detection of bacteremia in patients with sepsis, and may therefore be a pivotal step towards the clinical use of next-generation sequencing in this indication. To date, no next-generation sequencing-based test, also known as NGS-based tests for diagnosis of infectious diseases, have received pre-market approval or 510k clearance from the US FDA. However, several laboratory-developed NGS-based tests for pathogen detection directly from patient samples are available under the Clinical Laboratory Improvement Amendments, also known as CLIA certificates, at select commercial and reference labs. Now, there are different MNGS or metagenomic NGS testing available. There are three widely used NGS testing that I've mentioned here. The first one is the targeted NGS testing for pathogen detection directly from clinical samples, which is available from several reference labs. These tests are usually called universal or broad-range PCR testing, which begins with amplification of genes such as the 16S rRNA region for the bacteria, or the internal transcribed spacer, the ITS region for fungi using universal primers. Important thing to know about this test is that the Sanger sequencing or the tNGS does require samples from sterile sites, and the tissues when collected needs to be fresh and formalin-fixed and paraffin-embedded. The CARIUS test, the next one, which detects microbial cell-free DNA from blood plasma samples is actually among the most popular commercially available NGS test. This test has two major advantages, sample type and the turnaround time. First, the CARIUS test is performed on plasma, an abundant, easy to collect, and non-invasive sample, unlike the CSF or surgically collected samples. The use of plasma is based on the premise that during sepsis or serious infections, fragments of nucleic acids and the offending pathogens are shed into the bloodstream. Thus, the microbial cell-free DNA in the plasma can be a marker, not only for bloodstream infections or sepsis, but also other serious infections such as pneumonia, deep-seated abscesses, endocarditis, etc. Secondly, the CARIUS test has a stated turnaround time of two working days from the sample received. While there may be delays due to shipping, this is generally shorter than most culture-based or reference send-out lab test. The CARIUS test can detect more than 1,000 bacteria, fungi, parasites, and selected DNA viruses. The detected microorganisms are reported quantitatively as DNA molecules per microliter of plasma, aka MPM, and are compared to reference MPM ranges established in unhealthy asymptomatic individuals. The metagenomic next-generation sequencing, the MNGS pathogen diagnosis test, which is mainly available from the Department of Clinical Microbiology at the University of California, San Francisco, was first described clinical MNGS test for unbiased pathogen detection directly from patient samples. This test can detect bacteria, fungi, parasites, and rRNA and DNA viruses from multiple sample sites, including CSF samples, and are available to external clients. One thing that needs to be really cautiously used about the results of the MNGS test is that it can lead to unnecessary treatment or additional diagnostic investigations. The other thing that we need to be mindful about with this testing is that at more than $2,000 per test billed directly to the patient, the cost associated with the MNGS should not be overlooked, especially since this test is additive to current standard of care diagnostics. As such, the indiscriminate use of MNGS coupled with low rates of clinically significant or actionable results can lead to a relatively low return on investment. To summarize our talk, cultures are not dead yet. There's definitely ongoing need for phenotypic resistant testing. There are early antimicrobial therapy for targeted organisms can be administered, especially in presence of early identifications of resistant genes. While there are different modalities available to help rapid pathogen diagnosis, there are limited ability to distinguish between some important pathogens. All the system of rapid pathogen identification and diagnosis needs close coordination with the stewardship programs, and should be done in consultation with treating physicians, infectious diseases specialists, and clinical microbiologists to determine its appropriate use and interpretation. Overall, the rapid identification of pathogens can improve antimicrobial selection, it can reduce the toxicity of the antimicrobials administered, and can improve further outcomes. Thank you everyone for your time and the opportunity to speak to all of you today. I would like to especially thank Dr. Ryan Maves for his immense support and for his mentorship. Thank you once again. you

rapid diagnostics

pathogen identification

ICU

antimicrobial resistance

diagnostic methods

drug resistance