NEW YORK (GenomeWeb) – A recent study has extrapolated results from national hospital surveillance and laboratory surveys to determine the burden of Clostridium difficile infection in the US. According to the study, there were approximately half a million infections in 2011, resulting in nearly 30,000 deaths.
The reported infection estimate factors in molecular diagnostics use prevalence of 52 percent. This figure was arrived at from a survey of 37 labs, but the study, published last week at the New England Journal of Medicine, noted that national burden estimates are proportional to molecular test adoption, and would be about 200,000 cases higher with total adoption of molecular methods versus immunoassays.
The study also noted that about 66 percent of cases were associated with interaction with healthcare, while only 24 percent had onset during hospitalization.
"It's a surprising paper from the standpoint of identifying the major patient populations who need to be tested for C. diff," David Persing, executive vice president and chief medical and technology officer at MDx maker Cepheid, told GenomeWeb in an interview this week.
"It's not just inpatients, it is patients with a history of hospitalization and healthcare contact, patients with a history of antibiotic use," he said.
The fact that the majority of C. diff cases are not inpatients has been mentioned in the past, Persing said, but he added that he had not seen a number as high as 76 percent.
The study also reported that 82 percent of patients with community-associated C. diff said they had visited outpatient healthcare settings, such as doctors or dentists offices, in the 12 weeks before collection of C. diff positive stool samples.
"I take that to mean that they probably got antibiotics in that setting and that put them at risk for C. diff," Persing said. "That could be somebody who was a carrier of C. diff who got antibiotics, and that allowed the C. diff to overgrow," he said.
This in turn highlights an interesting subject of ongoing research. "Maybe there's more to just being a carrier than being a silent shedder of C. diff," Persing said, in which case a sensitive test may also predict susceptibility if they are prescribed antibiotics. Â
The NEJM study was conducted by researchers at the US Centers for Disease Control and Prevention. The feedback of testing method on estimated incidence rate was in line with a previous study, Fernanda Lessa, a medical epidemiologist at the CDC and first author on the NEJM study, told GenomeWeb in an interview.
Based on surveillance data from three US states, an earlier study published in Clinical Infectious Diseases reported that switching from toxin enzyme immunoassays, or EIAs, to nucleic acid amplification tests, or NAATs, for detection resulted in an increased incidence of C. difficile ranging from 43 to 67 percent.
For the NEJM study, Lessa and her colleagues projected the data on NAAT use in surveillance sites because national data are not available. The surveillance tracks details of the brands of tests being used, but the CDC does not report that information except to differentiate NAAT from EIA use.
"I can tell you right away that the most common one, and I think that probably is publically available information, is the Cepheid test," she said.
Indeed Cepheid is the leader in MDx sales for this infection, with 63 percent of the account shares for the US C. diff testing market, according to a presentation by the firm at the JP Morgan Chase Healthcare Conference last month.
Cepheid's cartridge-based Xpert platform, which runs the tests, is also "by far the most commonly used diagnostic platform," Persing said. "The reason we're adopted more widely is, it is just much easier to use, it's self-contained, there's not a contamination risk for the lab ... and it's accurate," he said. The firm also now manufactures an additional assay for an emerging highly-pathogenic strain of the bacteria, he added.
Cases of C. difficile are reported to the National Healthcare Surveillance Network. In 2017, these rates will also be part of the Center for Medicare and Medicaid Services value-based purchasing program, or pay-for-performance. In a telebriefing last week describing the NEJM study, CDC's Clifford McDonald said, "If you use a very sensitive test, but are very selective in how you decide it should be used or who to test, it's a very good way to diagnose diseases in general. If you use highly sensitive tests indiscriminately, you'll end up overdiagnosing the infections."
"When we report data on hospital rates, we take into account the type of test they're using so that, in the case of hospitals that are trying to perform better, they're not unduly penalized for using a more sensitive test," he added. "We're cognizant that over time we may see some countermovement; that cases might just go up because you're using more sensitive tests when you're really doing a better job in preventing the disease. And so we're taking that into account," McDonald said during the briefing.
Lessa confirmed this. "We know that the use of NAAT will increase your C. difficile incidence rate, but when we release the report of the hospital rates, we are controlling for the use of the NAAT," she said.
"Whenever we look at changes over time, [we] control for that. Every single hospital that is reporting to the National Healthcare Safety Network needs to fill out a survey at the beginning of every year telling the NHSN system which type of assay they are using for C. difficile diagnosis." said Lessa.
C. difficile bacterial infection produces a potent toxin which can lead to "deadly diarrhea," according to the CDC telebriefing. Patients may require partial or total removal of the colon if it is destroyed by C. diff toxin, and those who recover can be subjected to recurrences of the illness.
The bacteria are shed in spores which act like "ballistically hardened DNA delivery vehicles," Persing said. These spores are not easy to destroy by hand washing or standard cleaning protocols. They can survive on surfaces for long periods of time, thus infection control of patients who are shedding spores includes isolation. Cepheid, meanwhile, got into the molecular diagnostics business detecting anthrax spores, according to Persing. "Spores are us," he said.
It is thought that a balanced gastrointestinal microbiome can tolerate C. diff bacteria, but antibiotics can tip the balance and lead to a C. diff bloom. Not surprisingly, then, the infection is more common in people recently prescribed antibiotics. It is also considered a hospital-acquired infection.
In a hospital setting, deciding whom to isolate, as well as when a patient can be freed from isolation, also motivates the use of rapid and sensitive molecular testing. Both Lessa and Persing noted that there have been issues in the past of clinicians not trusting negative EIA results, which has in turn shaped clinical practice.
Lessa noted a study presented at a recent ID Week conference that examined termination of contact precautions. That project concluded that hospitals could ultimately save money by adopting NAAT tests, even though they are more expensive on a per-test basis, because a negative test is so much more reliable. In the study, "If the NAAT result was negative, the patient would be removed from contact precautions, as opposed to the EIA being negative [in which case] the patient continued on contact precautions because the physicians probably did not trust [the result]," Lessa explained.
The nuances of high sensitivity
However, there is also debate around test sensitivity, since theoretically C. diff DNA tests will detect the presence of the bacterium regardless of whether it is causing illness.
According to Persing, this debate is fueled by the competition in the diagnostics space, particularly companies that market immunoassays, which might say that molecular tests like Cepheid's are too sensitive while theirs are "not too sensitive, not too low sensitivity, just right in the middle," he said.Â
He pointed out that the vast majority of diarrhea cases tested in the hospital are negative for C. diff. "What you really need in that setting is a test that essentially rules out C. diff with confidence," he said.
On the basis of a single specimen, molecular testing can rule out C. diff diagnosis in 90 percent of samples, he said.
Diagnosis of diarrhea caused by C. diff, meanwhile, requires other clinical factors, and there could be an argument over whether a positive molecular test is discovering co-infection or colonization that is not causing the symptoms.
"At least you have the option now of considering whether they have diarrhea caused by C. diff, and all of those cases ― those 10 or 15 percent of cases ― ought to be in isolation. … Every one of those patients with diarrhea, if they have C. diff spores in their stool, is an infection control risk," Persing said, and contact precautions are needed to manage this biohazard situation irrespective of whether the diarrhea is caused by C. difficile toxin.
This is backed up by a Journal of Clinical Microbiology study which showed about 15 percent of samples wind up testing positive, Lessa said. Interestingly, that study also found that, "Once the hospital transitions over to a more sensitive diagnostic test, those hospitals implement a much more strict stool rejection policy … [which] leads to a decrease in the number of stools that they are testing. So the positivity rate goes up but the volume of stools being tested for C. difficile goes down because they are using a more sensitive test," Lessa said. The study further showed a reduction in repeat testing within a 48 hour period if hospitals are using NAATs versus EIAs.
Persing was aware of this ephemeron. "It's a one and done concept. There has been a serious trust factor among clinicians ordering [immunoassay] C. diff tests in the past. … Now [molecular tests] are so sensitive they don't need to order multiple tests," he said.
So, an increasing adoption of molecular tests drives up estimates of the disease burden. Yet hospitals may be using fewer tests overall and molecular testing could change clinical practice and ultimately drive down disease incidence. What might this mean for diagnostics test sales?
"We think the appropriate place for the technology is … to rule out the majority of C. diff cases confidently, to place patients into appropriate barrier precautions as quickly as possible to reduce spread of C. diff, and then to use molecular testing, ancillary testing, [and] clinical history data to ultimately make the diagnosis of C. diff diarrhea, which is not one that can be made based on stool testing alone," Persing said.
"We think, in that setting, the most sensitive technology is the appropriate one to use. … It is going to isolate patients appropriately, it's not going to over-isolate, it's not going to under-isolate, [and] it is going to save money for the lab because they don't have to run multiple tests per patient to rule out C. diff because they have one test that can confidently rule it out."
The CDC briefing cited the example of C. diff in the UK, where improved infection control and antibiotic stewardship ultimately contributed to a marked decrease in incidence, even considering adoption of NAATs.
"In order to see C. diff rates actually come down, you need a very sensitive test to be able to identify the cases in your hospital that are sources of infection, and you need appropriate infection control practices in place," Persing also explained. "If you systematically ignore shedders of C. diff in your hospital environment because your test is just not able to detect them, then you're not going to have those people in isolation, you're not going to be able to get good control over C. diff," he said, citing a 2013 study in the The Joint Commission Journal on Quality and Patient Safety.
"I would think that if CMS is going to put some teeth into this and expect hospitals to actually come down in their C. diff rates, that they're going to follow a protocol that's similar, which to me is one of the most effective demonstrations of the ability to reduce C. diff burden in a hospital environment," Persing said.
"In a sense the goal of every infection control program is to put yourself out of business, to eliminate the need for your services anymore because you eliminate all source of infection," he noted. "Declining C. diff rates in the US would be the best thing we could ever see as the result of our technology," he added.
Persing noted that this would likely not affect Cepheid's business because most C. diff testing is done on the suspicion of C. diff.
From the CDC's perspective, molecular testing is an important improvement, Lessa said.
"We look at public health, we look at the safety of the patient, we look at clinical management, and of course we want tests that are rapid, more sensitive, and easy to be done in the laboratory," she said. "I think that NAATs offer those characteristics, [they are] better for the clinical management of the patient because you have the diagnosis that's very rapid, and they are better for infection control."Â