Culture clash: New diagnostics complicate public health surveillance
Advanced diagnostic technologies allow clinicians to promptly initiate appropriate treatment and improve patient outcomes. Efficient, highly sensitive molecular and antigen-based tests have become widely adopted by clinical and commercial laboratories across the United States, according to the CDC, presenting a significant hurdle for long-established disease surveillance programs that rely on clinical cultures to track infectious diseases.
“That sample we receive from the hospitals and clinics is used to fully characterize the bacteria or virus so that we can determine whether it can be readily treated with antibiotics,” John M. Besser, PhD, deputy chief of the Enteric Diseases Laboratory Branch at the CDC, told Infectious Disease News. “We also use this material to look for types of trends [and] detect outbreaks, so the public health system needs to adapt in a variety of ways to this new type of information.”
Infectious Disease News spoke with public health and clinical laboratory experts about the benefits of culture-independent diagnostic tests (CIDTs) and ongoing efforts to incorporate this technology into public health surveillance methodology.
Improved speed, sensitivity boost adoption of CIDTs
In a 2015 report published in Emerging Infectious Diseases, Gayle E. Langley, MD, medical epidemiologist in the CDC’s Respiratory Diseases Branch and medical director of the Active Bacterial Core surveillance program, and colleagues detailed the shift toward greater use of CIDTs. Referencing data collected through the CDC’s Emerging Infections Program (EIP) from 2011 to 2014, they wrote that all or nearly all of the Clostridium difficile, Legionella spp., Bordetella pertussis and influenza cases reported through the system were diagnosed by either molecular or antigen-based tests.
The use of CIDTs for enteric infections such as Campylobacter and Shiga toxin-producing Escherichia coli is also becoming more common. Findings published in MMWR by Martha Iwamoto, MD, of the CDC’s Division of Foodborne, Waterborne and Environmental Diseases, and colleagues suggested that 46% of the 5,614 positive CIDTs reported to the Foodborne Diseases Active Surveillance Network (FoodNet) from 2012 to 2014 were not confirmed by culture. Preliminary data recently reported by Jennifer Y. Huang, MPH, also of the CDC’s Division of Foodborne, Waterborne and Environmental Diseases, indicated this rate increased to approximately 60% in 2015.
“It’s happening fairly frequently that specimens are not cultured,” Besser said. “At this point in time, the culture-independent tests still make up a minority of overall testing that is done in the country. But we expect that to change in the next few years.”
Such practices are markedly different than those employed in previous decades, when diagnoses were primarily reliant on culture, according to Angela M. Caliendo, MD, PhD, professor and vice chair of medicine at Alpert Medical School of Brown University.
“Look back at the mid-‘90s when we all started thinking molecular — most clinical virology labs did quite an extensive array of culture, and flu isolates went to the CDC so that they could figure out how to manufacture the vaccine every year,” Caliendo told Infectious Disease News. “You fast forward over 20 years and now there’s very little culture being done in clinical virology labs. There’s still some, but nowhere near [what it was] — the whole field has been transformed.”
Early CIDTs invigorated clinical virology by reducing the time and expertise needed to process a viral sample, Caliendo said, and they were more often sensitive than culture and allowed quantification of the amount of virus in a given sample. Simultaneously, multiple highly sensitive molecular tests for bacteria such as Neisseria gonorrhoeae were approved by the FDA and hailed by researchers as a flexible and efficient alternative for settings where optimization of culture would be difficult.
Although several new tests continued to be developed for other “low-hanging fruit,” Caliendo said the demand for faster and more efficient diagnostics for respiratory viruses skyrocketed after the 2009 H1N1 influenza pandemic.
“Many labs were still culturing respiratory viruses at that point,” Caliendo said. “With that outbreak, it became clear that our antigen tests weren’t very sensitive. There was a need for rapid molecular diagnostics, and the number of tests has since exploded.”
Most of the molecular tests being adopted today are highly sensitive, Langley and colleagues wrote, especially for bacterial cases in which previously administered antimicrobials could inhibit culturing. Of equal import is their speed, which often is the deciding factor in clinical practice when treatment is urgently needed, according to Adriana Weinberg, MD, professor of pediatrics, medicine and pathology at the University of Colorado-Denver Anschutz Medical Center and medical director of the clinical molecular and virology laboratories at the University of Colorado Hospital.
“[Nucleic acid tests] are more sensitive than culture,” Weinberg told Infectious Disease News. “One can issue a result faster, which is important, and also these tests can pick up minor quantities of the pathogen, which culture does not have the ability to do. From the standpoint of patient care, these tests are advantageous.”
Many of the CIDTs approved by the FDA in recent years are multiplex panels able to test for multiple infections at once. While Besser noted that these panels offer savings in cost and manpower, Infectious Disease News Editorial Board member Timothy F. Jones, MD, state epidemiologist for the Tennessee Department of Health, said the adoption of multiplex CIDTs also has led to many unexpected diagnoses.
“It used to be that you had to ask for a routine stool culture, for ova and parasite tests, or you had to request viral stool testing, which is specific and not readily available,” Jones said during an interview. “Now, you’ve got these panels which can test for 22 different things, including parasites, bacteria and viruses. It’s really handy, and we’re starting to find a lot of things that people were never looking for before.”
One such example is sapovirus, an enteric virus that, according to a study of viral gastroenteritis outbreaks in long-term care facilities from 2002 to 2009, was often missed or diagnosed as norovirus before being detected with reverse transcriptase-PCR. Similarly, Jones said the detection of polymicrobial infections not initially suspected by a treating physician has been enhanced by multiplex panels as well, improving outcomes for patients whose infections would otherwise have avoided notice.
Increased detection complicates diagnoses, disease burden estimates
The appeal of CIDTs’ increased sensitivity and speed is not exclusive to clinical practice, Besser said, as public health also has plenty to gain from the technology — provided these new practices can be incorporated into existing surveillance methodology.
“The tests have some potential benefits for public health in that we can potentially get information about illness in the general public more quickly, just as the physicians are getting their information more quickly,” Besser said. “But, it presents a certain set of problems for public health that we need to adapt to.”
Among these are the various factors associated with CIDTs that could have a sizable effect on EIP’s disease burden estimates, Langley and colleagues wrote. An increase in the prevalence of a disease could be the result of improved sensitivity, as opposed to an actual increase in incidence. Conversely, specificity data from a falsely identified outbreak of B. pertussis detailed in Pediatrics indicated the possibility that false positives could inflate estimates. The ease and flexibility of multiplex panels could lead to changes in clinical testing practices while the clinical definition of a positive case is rewritten with each new diagnostic, both of which could substantially impact reports of disease burden.
If estimates are to remain stable, these concerns must be considered as the use of CIDTs continues to increase, Huang and colleagues wrote.
“The ability to assess and interpret change is impeded as the number of positive CIDT reports continues to rise because of important limitations in the understanding of CIDTs and possible changes in clinician and laboratory practices surrounding them,” they wrote. “Surveillance systems need to adapt to these changes by expanding case definitions to include positive CIDT reports.”
CIDTs hamper pathogen characterization
More concerning, according to Besser and Jones, is the dearth of clinical isolates developed from cultured specimens, which are critical for the microbiological and molecular characterizations currently conducted by public health.
“The public health community is behind in catching up with this new technology,” Jones said. “They’re very concerned, because a lot of the traditional molecular fingerprinting and things that the public health folks depend on require cultured specimens.”
Because CIDTs do not result in an isolate, many of the surveillance networks established by the CDC — such as EIP, FoodNet and PulseNet — face a potential shortage of new data. These programs, Langley and colleagues wrote, provide critical services in vaccine development, treatment evaluation and outbreak identification, all of which could be lost or hampered by the use of CIDTs. Furthermore, many emerging strains, mutations or antimicrobial drug resistances are discovered by public health researchers with access to isolates, and CIDTs run by clinical labs are only capable of testing for previously identified phenotypic resistances.
“You can do PCR or these other tests and determine that the organism is there, and you can actually do PCR for many of the known genes that cause antibiotic resistance,” Jones said. “The problem is you could only test for things that you already know about. When we had the culture, you can throw antibiotics on it, see if it dies — if it didn’t, you had your clinical answer. If a bug were to mutate, and the new gene popped up that causes antibiotic resistance, we wouldn’t know because we’re testing for genes, not for the antibiotic resistance itself.”
Taken together, these issues could delay detection of various severe or resistant outbreaks unless surveillance networks — such as PulseNet — are able to adapt.
“Many of the outbreaks that one hears about in Salmonella and Listeria and [Shiga toxin-producing E. coli] are detected by PulseNet — this flow of bacteria coming from clinical health laboratories to public health laboratories,” Besser said. “These outbreaks represent problems in the food supply that can be corrected if we recognize that the problems exist. ... That’s what we’re trying to protect by making a measured response to this problem of CIDTs.”
Analysis of specimens relies on cooperation
Previously, the problems related to pathogen characterization could have been mitigated by requesting that clinical and commercial labs perform reflexive cultures after a positive CIDT result. While this practice may have been possible when diagnoses still required culture confirmation, Jones said the growing reliance on CIDTs has made this option unfeasible.
“It’s unreasonable for public health to demand that a commercial lab go back and do additional testing because they’re doing CIDTs to save time and money — it’s not their business,” he said. “The CDC tried to pull that off for several years and labs were remarkably cooperative, but that’s just becoming a more and more unreasonable request.”
Along with cost concerns, Weinberg stressed that the role of hospital or commercial labs is to facilitate clinical diagnoses, not discover new resistance mutations. However, she said that most health care professionals understand the importance of surveillance and would be willing to work with public health to develop new, unobtrusive procedures that will permit culture isolation of pathogens.
“Public health needs to invest in performing their own cultures, or partnering with clinical laboratories to perform cultures, which would solve the important problem of detecting new resistance,” Weinberg said. “[The stance] of most hospitals and my own clinical lab has always been to provide the CDC and [state health departments] the specimens that they need to do their public health work — it’s very important, and we’ve always supported it. We all populate the same Earth, we’re all interested in having good, strong public health.”
David Boxrud, MS, molecular epidemiologist supervisor at the Minnesota Department of Health, agreed. Although much of the burden continues to fall onto the shoulders of public health labs, many of the clinical labs within his department’s jurisdiction and others have largely been receptive to outside requests for specimens.
“[We] have had a lot of discussion with clinical labs about what we would prefer to get and making sure that at least we get specimens, because without specimens our whole surveillance system is down,” Boxrud told Infectious Disease News. “Nationally, there’s been a lot of discussion with testing companies to make them aware of the issue and try to write their protocols in such a way that public health can still perform their work.”
Many of these discussions have been spearheaded by the CDC and professional organizations, according to Boxrud. Following a national meeting of government health officials, public health epidemiologists and representatives from clinical and commercial labs, committees from the Association of Public Health Laboratories (APHL) and the American Society for Microbiology (ASM) recently collaborated to release interim recommendations for clinical microbiologists who wish to submit enteric CIDT-positive specimens for public health testing. These guidelines, which will be updated after analysis of isolate recovery studies, recommend that clinical labs:
- contact local or state public health to clarify specific regulations or preferences concerning specimen provision prior to CIDT implementation;
- continue to obtain isolates when pathogens requiring submission are detected, if feasible; and
- submit CIDT-positive specimens such as stool or gram-negative broth to a public health lab within 24 hours of the CIDT result, should isolate culture be unavailable.
According to the APHL, “Adhering to these Interim Guidelines for Submission of Positive CIDT Specimens to Public Health will lead to ... accurate pathogen surveillance to monitor trends in disease, timely identification of outbreaks caused by enteric pathogens, removal of contaminated products from commerce and the ability to monitor industry changes designed to improve the safety of the food supply.”
Sentinel programs, new technologies offer long-term solutions
While obtaining nonculture specimens from clinical and commercial labs provides a tentative solution, Boxrud said more lasting solutions will soon be needed. While Boxrud noted the possibility of regulatory measures to promote isolate production, Jones pointed to the long-running Gonococcal Isolate Surveillance Project as another option.
“Should certain sentinel labs continue to grow the organisms and do traditional culture, just so we don’t completely lose that capacity?” Jones asked. “[In 1986] they actually set up a very specific national network of sentinel laboratories ... no one else was going to catch any of these mutations because the early CIDTs were so good. That’s the kind of thing we may have to think about in the future.”
The movement away from culture dependence itself could offer a long-term answer to the problem, as well. Langley and colleagues hypothesized that whole-genome sequence libraries and metagenomics could aid both specimen characterization and public health surveillance, but noted these options “pose challenges for processing, analyzing and interpreting large amounts of data.”
On the other hand, Besser, Boxrud and Caliendo each noted that ongoing research into sequencing technologies is promising, and described diagnostics now in development that are capable of obtaining microbiological and advanced pathogen characterization data directly from a clinical specimen.
“In the short term, we need to make sure that we continue to get isolates ... so that we can continue to characterize them,” Besser said. “In the long term, we need to develop new methods for public health that are themselves culture-independent, so that we can rapidly characterize these specimens to a high level, and we can get the information that we need quickly — just as the physicians are getting information quickly. [These advancements] will require a certain amount of development.” – by Dave Muoio
- APHL. Submission of Enteric Pathogens from Positive Culture-Independent Diagnostic Test Specimens to Public Health. http://www.aphl.org/AboutAPHL/publications/Documents/FS-Enteric_Pathogens_Guidelines_0216.pdf. Accessed April 13, 2016.
- CDC. Gonococcal Isolate Surveillance Project. http://www.cdc.gov/std/gisp. Accessed April 13, 2016.
- FDA. Nucleic Acid Based Tests. http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitroDiagnostics/ucm330711.htm. Accessed April 13, 2016.
- Huang JY, et al. MMWR Morb Mortal Wkly Rep. 2016;doi:10.15585/mmwr.mm6514a2.
- Iwamoto M, et al. MMWR Morb Mortal Wkly Rep. 2015;64:252-257.
- Koumans EH, et al. Clin Infect Dis. 1998;doi:10.1086/514994.
- Langley G, et al. Emerg Infect Dis. 2015;doi:10.3201/eid2109.150570.
- Lee LE, et al. Emerg Infect Dis. 2012;doi:10.3201/eid1805.111843.
- For more information:
- John M. Besser, PhD, can be reached at firstname.lastname@example.org.
- David Boxrud, MS, can be reached at email@example.com.
- Angela M. Caliendo, MD, PhD, can be reached at firstname.lastname@example.org.
- Timothy F. Jones, MD, can be reached at email@example.com.
Disclosures: Weinberg reports research relationships with BD Becton Dickinson and Roche. Besser, Boxrud, Caliendo, Huang, Iwamoto, Jones, Koumans, Langley and Lee report no relevant financial disclosures.