Feature Story

Two more for molecular methods: CNS, GBS

January 2004

William Check, PhD

Second of two parts on using molecular methods to detect viral and bacterial pathogens. Last month: diagnosing respiratory syncytial virus and pertussis. This month: infections of the central nervous system and group B streptococcus.

Molecular diagnostics play a major role in identifying the etiology of infections of the central nervous system, particularly aseptic meningitis, says Yi-Wei Tang, MD, PhD, associate professor of medicine and pathology and director of the molecular infectious diseases laboratory at Vanderbilt University Medical Center, Nashville. Such infections are relatively common among children, especially in summer, manifesting as headache and vomiting. Worried parents bring their children to the emergency room and, because the differential diagnosis includes potentially lethal organisms such as herpes simplex virus and signs and symptoms are similar for mild and severe pathogens, children may be hospitalized and given combination therapy such as acyclovir, gentamycin, and ampicillin.

In this circumstance, organism-specific diagnosis of CNS infections would be clinically helpful. A positive test for enterovirus, which typically has a mild course, combined with a reliable negative test for HSV, could allow children to be sent home up to two days earlier without treatment (but with close followup), saving money and avoiding unnecessary use of antibiotics.

PCR-based molecular methods have already become the gold standard for detecting HSV infection in the cerebrospinal fluid of patients with encephalitis and meningitis, Dr. Tang says. However, it is common to perform post-amplification confirmation with lengthy and time-consuming procedures such as probe hybridization by Southern blotting, which reduces the potential benefit of PCR. While working as a fellow at the Mayo Clinic, under the supervision of David Persing, MD, PhD, Dr. Tang developed a colorimetric microtiter plate method that confirms the diagnosis of HSV much faster than Southern blotting with no loss in sensitivity or specificity (Tang YW, et al. J Clin Microbiol. 1998; 36: 2714-2717).

In Dr. Tang’s microtiter plate procedure, digoxigenin-labeled PCR amplicons are made by including digoxigenin-dUTP in the amplification step. A biotin-labeled oligonucleotide probe is added to the denatured amplification product to capture labeled amplicons. Probe-amplicon hybrids are then immobilized on a streptavidin-coated microtiter plate, and an antibody-enzyme conjugate is added to produce color, visualizing the virus being probed for. Dr. Tang says this method produces results in 12 to 24 hours less than Southern blotting and has the additional advantages of being user-friendly, controlling carryover contamination (because the incorporated dUTP makes the reaction mixture sensitive to enzyme treatment), and providing an additional 10- to 100-fold signal amplification, thereby increasing sensitivity.

He and his colleagues evaluated this method versus Southern blotting using three commercial microtiter plate systems and the Mayo system. Specimens were 86 cerebrospinal fluid samples submitted for HSV diagnosis by PCR, of which 54 were positive by Southern blotting. Two of the three commercial microtiter plate systems and the Mayo system performed as well as Southern blotting, with sensitivities of 98 to 100 percent and specificities of 97 to 100 percent. Turnaround for the microtiter plate system was less than eight hours including specimen processing and PCR amplification, compared with more than 24 hours for Southern blotting hybridization.

Further potential savings from PCR can be realized by avoiding unnecessary ordering of multiple viral tests on CSF specimens submitted for HSV testing. Clinicians, particularly new house officers, often order HSV 1 and 2, cytomegalovirus, Epstein-Barr virus, varicella-zoster virus, and human herpesvirus 6 and 7 on the same sample. Dr. Tang and his colleagues at Mayo reviewed test results for three years, asking whether initial screening of CSF samples by leukocyte counts and protein levels would help. They found no HSV positives among specimens normal for both parameters, a one to three percent positive rate for samples with one abnormal measure, and a 30 percent positive rate for HSV when both criteria were abnormal (Tang YW, et al. Clin Infect Dis. 1999; 29: 803-806).

Dr. Tang concludes that molecular detection of HSV DNA in cerebrospinal fluid, now considered the gold standard, acquires enhanced sensitivity and specificity with more rapid turnaround when a "DNA EIA" using a microtiter plate method is employed for additional product identification after PCR. Eliminating samples submitted for HSV PCR that do not have both abnormal leukocyte count and protein concentration saves money without adversely affecting patient care, he says. Further gains can be realized by reducing multiple ordering.

For preventing neonatal infection with group B streptococcus, improved molecular methodology not only enhances detection but also goes hand-in-hand with a more effective public health policy. Infection with GBS, the popular name for the hemolytic gram-positive organism Streptococcus agalactiae, is a leading cause of neonatal morbidity and mortality, explains Michel G. Bergeron, MD, professor and chair of the Division of Microbiology and Infectious Disease Research Center at Laval University, Quebec City. Ten to 30 percent of pregnant women have GBS colonization of the gastrointestinal or genital tract, with vertical transmission in many infants born to these mothers. As a result, one to three percent of infants get GBS neonatal disease.

Before therapy, the case-fatality ratio was as high as 50 percent. Guidelines issued in 1996 recommended intrapartum antibiotic prophylaxis after screening at 35 to 37 weeks of gestation. Antibiotics are administered if a woman is culture-positive or has one or more risk factors: fever during labor, membrane rupture 18 hours or more before delivery, or delivery before 37 weeks’ gestation. Screening decreased the incidence of early-onset GBS disease (<= seven days after birth) from 1.7 cases per 1,000 live births in 1992 to 0.6 cases in 1999. Still, GBS remains a leading cause of serious neonatal infection.

A cohort study comparing the two screening approaches among more than 600,000 live births in the United States found that culture-based screening is at least 50 percent more effective than risk-based appraisal in preventing early-onset GBS (Schrag SJ, et al. N Engl J Med. 2002; 347: 233-239). As a result, prevention guidelines were revised in 2002 to recommend universal prenatal screening with culture at 35 to 37 weeks of gestation (Schrag S, et al. MMWR Recomm Rep. 2002; 51[RR-11]: 1-22).

However, current methods for identifying GBS are less than ideal, Dr. Bergeron says. Culture takes too long and direct immunological tests are only about 60 percent sensitive. Phenotypic tests, even automated ones, are also complex, requiring 10 to 25 steps, and take one to three days. Existing genotypic tests, while more sensitive, also have many steps and take several hours to one day to provide results.

Dr. Bergeron set out to develop a rapid real-time PCR assay for GBS. By combining rapid (10-minute) sample preparation and DNA extraction followed by real-time PCR, he succeeded in devising an assay with a turnaround time of less than one hour. When the assay was evaluated in 112 parturient women, both the novel assay and conventional PCR achieved a positive predictive value of 100 percent for GBS and negative predictive value of 98.8 percent compared with culture. Time to acquire results was 30 to 45 minutes for the novel PCR assay, 100 minutes for conventional PCR, and more than 36 hours for culture (Bergeron MG, et al. N Engl J Med. 2000; 343: 175-179).

An unpublished prospective multicenter study sponsored by Infectio Diagnostic, the company that manufactures and markets a kit based on Dr. Bergeron’s method (IDI-Strep B), found that this method could greatly improve clinical sensitivity of antepartum screening relative to culture with no loss in clinical specificity. "This assay, which is approved by the FDA, can be used in place of standard culture methods," Dr. Bergeron says.

He says this novel IDI-Strep B test, which is now sold in Canada and the United States, is so rapid that it allows screening while a woman is in labor. "Delivery is the optimal time for screening," Dr. Bergeron says. To take advantage of the test’s rapid turnaround time, it must be available on demand. Because of its simplicity-only six steps-Dr. Bergeron says it is feasible to have chemistry staff perform it, making around-the-clock coverage possible. "As new and more rapid DNA-based tests become available, the microbiology laboratory will be accessible 24 hours a day," he says.

Dr. Bergeron’s research group is developing other assays using this innovative real-time PCR method. The next one expected to appear on the market will be for rapid (under one hour) detection of methicillin-resistant Staphylococcus aureus. "Stat DNA-based microbiology is here," he says. "It will revolutionize the practice of medicine." In Dr. Bergeron’s perception, some logistical hurdles remain to eliminating empirical infectious disease practice by the widespread adoption of molecular methods for identifying pathogens directly. "Practicing without culture will require a change in culture," he says.

Christine Ginocchio, PhD, director of microbiology/virology and molecular diagnostics, North Shore-Long Island Jewish Health System Laboratories, acknowledges that some assays are simple, with quick extraction, cookbook recipes, and easy insertion of samples in an instrument. She cites Becton Dickinson’s ProbeTec for chlamydia/gonococcus as one example. However, she says, "Putting all molecular assays in the same technical category and having anyone rotate in from the general laboratory staff to do real-time PCR is somewhat of an oversimplification. We who are trained in molecular microbiology cringe when we hear that every molecular test can be run by anyone in the laboratory. Someday the assays may all be on an automation level comparable with that of chemistry, but unfortunately, the majority of them today are not."

She wonders if general staff will be able to troubleshoot or even recognize a problem when it occurs if they are not familiar with the science of the methods and the parameters and limitations of molecular-based testing. Importantly, interpreting data is not always straightforward. "If a CSF sample is positive for HSV 1, that’s all you need to know," she says. But if the CSF of an HIV-infected patient is positive for CMV, that may mean nothing clinically.

"Molecular technologists are often asked highly technical questions by the medical staff, and the ability to explain the assays and what the results could mean is often critical for the physician to make a correct interpretation," Dr. Ginocchio says. "To run these assays, technologists must not only have the skills and experience, but also like to do this type of testing and have an appreciation for fine detail and adherence to protocols."

William Check is a medical writer in Wilmette, Ill.