College of American Pathologists
Printable Version

  Antibiotic program supersizes impact of molecular testing


CAP Today




October 2011
Feature Story

Anne Paxton

When Frederick V. Plapp, MD, PhD, addressed the Compass Group conference on best practices in microbiology and infectious diseases last May, some of the laboratory leaders in the audience had never heard of the concept of “antibiotic stewardship.” But if Dr. Plapp gets his wish, they’ll be hearing far more about it in the future.

As he demonstrated, in combination with rapid PCR and PNA FISH tests for infectious agents, antibiotic stewardship can significantly curtail unnecessary hospital admissions and improve antibiotic administration. In the case of Saint Luke’s Regional Laboratories in Kansas City, Mo., where Dr. Plapp is medical director, a small pilot study demonstrated that stewardship could slash the average number of antibiotic dosages per patient by more than half, from 5.8 to 2.8. “I’d like to see every hospital adopt an antibiotic stewardship program,” he said recently in an interview with CAP TODAY.

The Centers for Disease Control and Prevention defines such programs as interventions designed to ensure that hospitalized patients receive the right antibiotic at the right dose, for the right duration and at the right time. Dr. Plapp describes two initiatives at Saint Luke’s that show the value of not only reporting infectious disease laboratory results fast but also intervening to ensure the results are put to optimal use.

The first initiative was geared to the adoption of a new meningitis protocol. Tests to distinguish cases of viral meningitis from bacterial meningitis count heavily in determining hospital admission, length of stay, and antibiotic treatment of patients who present with symptoms of infection. Most cases of viral meningitis are caused by enteroviruses, from which the patient will usually recover completely after seven to 10 days, and hospital admission is not needed. For bacterial meningitis, the prognosis without treatment can be far more serious. If caught early, bacterial meningitis can be treated with a number of effective antibiotics.

Under the traditional diagnostic model, Dr. Plapp notes, if meningitis is suspected when a patient presents to the emergency room, the ER physician will order a lumbar puncture, blood cultures, and hospital admission pending the microbiology results. A cerebrospinal fluid bacterial culture requires three days while a CSF viral culture requires four to 10 days. Meanwhile, the admitted patient is treated with IV antibiotics until it’s likely that the bacterial culture will be negative.

Most hospitals still follow this model, Dr. Plapp says. But the emergence of PCR testing for enterovirus is bringing a new era, and Saint Luke’s wanted to take advantage of it. In 2004, “we decided to switch enterovirus testing away from culture, and we started using the Roche LightCycler and doing the test in our molecular department just at one of our hospitals.” The test was offered Monday through Saturday and had a 24-hour turnaround time.

Unfortunately, this new arrangement did not affect physicians’ admitting practices. “It really had no impact whatsoever,” Dr. Plapp says. Concluding that it wasn’t turning around the results fast enough, the laboratory took a different tack in a pilot study later that year.

During the peak enterovirus season in Kansas City, from July to October, the laboratory provided its molecular technologist with a pager and gave the number to the ER physicians at three of the system’s nine hospitals. “They would page our technologist and ask us to do a stat real-time PCR for that patient.” Once the specimen arrived, the lab tried to get the results to the physician within four hours, using a stat courier for two of the satellite hospitals.

When the result was positive, it led to patients being discharged from the ED within two hours of the time the result was reported. “What this told us was that we really had to be able to do this test on a stat basis so you’d get results back while the patient was still in the ER and they hadn’t made decisions yet on whether to admit.” The results of the stat enterovirus testing were so impressive that shortly after the pilot project, the senior vice president for the Saint Luke’s Health System had Dr. Plapp present the pilot’s results to insurance company CEOs and medical directors. “He was very positive and thought it was a really unique program that distinguishes Saint Luke’s from other health systems in the region, and would allow them to save money,” Dr. Plapp says.

Of course, there were additional costs to providing a test on a stat basis. “The technologist carrying the beeper has to drop everything else and get that specimen, do the testing, and call the ER physician; it really takes them away from their other work.” After the laboratories switched to Cepheid’s GeneXpert test in 2007, that problem eased. “While it’s probably 30 to 40 percent more expensive than the Light Cycler, GeneXpert is much easier technology to run and the testing is not as technically demanding, so you don’t have to have a person who’s fully trained in molecular diagnostics to do the testing. We could actually do the testing in our microbiology department, which is staffed 18 hours a day,” Dr. Plapp says.

The savings in hospital care is more than enough to recover the cost. In 2010, the laboratory had 122 orders for enterovirus PCR testing at a cost of about $24,000, Dr. Plapp says. The number of positive patients happened to be small that year—only six—but with the average hospital admission cost for a suspected case of meningitis about $4,995 for 2.7 days of care, “we figured out we had a cost avoidance of almost $30,000.”

With just five positive cases in a year, the laboratory calculates that the cost of the enterovirus PCR testing would be offset. As he explains, “We avoid admission if the virus is detected, the patient avoids being exposed to IV antibiotics, they avoid having outpatient antibiotics, and we avoid the additional diagnostic test and procedures that might be used to work that case up further after admission.”

Meningitis tends to be a disease that afflicts younger people, and Saint Luke’s patients are mostly adults. “If we were seeing more children, we’d have an even higher positive rate and would be able to send even more people home than we currently do.” Still, Dr. Plapp was surprised that only 20 percent of the results for adults were positive for enterovirus, he tells CAP TODAY. “I guess there are a lot of cases where it’s truly viral meningitis and yet we’re not detecting it. We don’t know if it’s just because there are very few viral particles there, and the test is not quite sensitive enough, or if there are other viruses that are responsible.”

The cost might be a barrier to some hospitals hoping to follow a similar meningitis protocol, he says. “All hospitals are equally likely to get meningitis cases,” but the smaller labs are less likely to have the needed equipment. Large urban hospitals are more likely to make the transition to PCR enterovirus testing, he believes.

The second initiative at Saint Luke’s originated when the microbiology laboratory, under the direction of Cynthia Essmyer, MD, started its antibiotic stewardship program in 2003, built around a close working relationship between laboratory and pharmacy. Twice a day, the la-boratory sent inpatient culture and sensitivity reports to the hospital PharmD in addition to sending them to the attending physician. “Then the PharmD would call up the attending physician and recommend antibiotic changes based on what the results were,” Dr. Plapp says. Using this approach, his laboratory was able to expedite changes in antibiotics within 18 hours of the time the patient was admitted.

It’s increasingly common to see clinicians working with clinical pharmacologists, he notes. “The pharmacologists are in the hospital all the time. And more and more doctors are having them do rounding, and on their service they handle the antibiotic dosage. When we were just reporting results to the doctors, the doses didn’t change much, but once we started reporting results to PharmDs, they’d go to the floors and see patients, and they would change the antibiotics.”

The laboratory hoped to improve its antibiotic stewardship even further through the use of AdvanDX’s PNA FISH (peptide nucleic acid in situ hybridization). PNA FISH is a nonamplified fluorescent molecular probe that recognizes the ribosomal RNA unique to each different bacteria species. “They’re able to tell different bacteria in a very quick fashion,” Dr. Plapp says, “so it allows us to do very timely and accurate same-day results for a bacterial identification once we detect growth. It has to incubate for an hour and a half, but we can print out a result in about six hours, where normally it would take two days.”

In January 2009, “We started out just doing Staphylococcus aureus and coagulase-negative staphylococci. We then added the gram-negative rod test in July of 2009,” he said. In April 2010, AdvanDX introduced a new test called Traffic Light that tests simultaneously for Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa. Saint Luke’s acquired this test, despite the relatively high cost. “We try to batch our testing as much as possible to keep our costs down. I don’t think a lot of labs have adopted PNA FISH, partly because of the expense,” Dr. Plapp says.

But the benefits of the program are striking. In 2010, of 24,400 blood cultures tested at the hospital, 27 per-cent were coagulase-negative staph, potentially blood culture contaminants for which no antibiotics are necessary. Nineteen percent were MRSA (methicillin-resistant Staphylococcus aureus), 11 percent were E. coli, three percent Klebsiella, and almost three percent Pseudomonas. “So 62 percent of all of our blood cultures were organisms that were covered by PNA FISH, and we were able to identify them the exact same day that we detected growth on the blood culture.”

By employing PNA FISH, Dr. Plapp’s laboratory was able to reduce the average time from start of culture until bacterial identification from 132 hours (with culture alone) to 38 hours.

The average antibiotic cost per patient plunged from $72 to $24. But even more improvement came from involving the PharmD. “When we were doing PNA FISH and just reporting the results, there was no change in intervention by the physicians. But when we started involving the pharmacist and enhancing our stewardship program, we cut the amount of antibiotic dosages in half.” Another benefit: Antibiotic therapy could be optimized much sooner.

Antibiotic stewardship is here to stay, in Dr. Plapp’s view. “I think it’s the future. There are 235 million doses of antibiotics per year prescribed in the U.S. and anywhere from 20 to 50 percent of those are probably unnecessary. Bacteria are becoming resistant to antibiotics, but if we can reduce unnecessary exposures, it’s less likely that patients will become infected with resistant bacteria. So I think we will be doing more and more of this type of work.”

Anne Paxton is a writer in Seattle.