Cover Story

Molecular testing can facilitate diagnosis of respiratory infections

Respiratory viral infections have the greatest disease burden in young children, high-risk adults and the elderly. Diagnosis of these infections can sometimes present a challenge, even with available rapid testing.

Infectious Disease News discussed with several experts the topic of respiratory infections and reliability of molecular testing.

“There aren’t magic bullets for these respiratory viruses and, unfortunately, they can be severe,” said Paula Revell, PhD, assistant professor in the department of pathology and pediatrics at Baylor College of Medicine.

New detection methods have taken a lot of the “guess work” away from clinicians when trying to diagnose a patient, according to Andrea T. Cruz, MD, MPH, assistant professor of pediatric emergency medicine at Baylor College of Medicine.

“In our hospital, you can order seasonal respiratory panels that use polymerase chain reaction to detect the most common viruses seen during different times of the year,” she said. “For example, the winter panel includes influenza A and B, parainfluenza viruses, RSV, adenovirus, and human metapneumovirus, while the summer panel includes enterovirus, adenovirus, and parainfluenza viruses. The nice thing about the PCR panel is that it is inordinately more sensitive than rapid assays and does not rely on physicians knowing the seasonality of viruses, which can vary demographically.”

Historically, physicians had to rely on clinical diagnosis and rapid tests, which Revell said were not reliable.

“The test historically performs with suboptimal sensitivity, and frankly, a good clinician is as good at diagnosing influenza in season as the rapid test,” she said. “In fact, a good clinician is usually better than the test.”

Although there are vaccines for some respiratory viruses, such as influenza, vaccines are not available for most of them. Revell said prevention of these viral infections is the “best medicine.”

Gregory A. Storch, MD, from Washington University School of Medicine, said that molecular testing can detect multiple respiratory pathogens.

Gregory A. Storch, MD, from
Washington University School of Medicine,
said that molecular testing can detect
multiple respiratory pathogens.

Photo courtesy of Storch GA

“We would emphasize things like good hand hygiene and teaching little kids to sneeze into their elbows,” she said. “Our best tool to prevent influenza, in particular, is the vaccine, so vaccinate your kids and vaccinate yourself. That’s the best thing that we can do.”

New detection methods

Several new commercial tests have become available using molecular technology to detect a number of different respiratory viruses at the same time.

“Rapid progress is being made in molecular testing, and this is an area that should be revisited at frequent intervals,” Infectious Disease News Editor Emeritus Theodore C. Eickhoff, MD, said.

According to Gregory A. Storch, MD, director of the division of pediatric infectious diseases at the Washington School of Medicine, St. Louis, there are four large multiplex tests that have been approved by the FDA — Luminex xTAG RVP, Luminex RVP FAST, BioFire FilmArray Respiratory Panel and GenMark eSensor Respiratory Viral Panel.

“The large multiplexes are assays that detect more than three viruses at the same time,” Storch said. “All four of the FDA-approved assays do well and are fairly similar to each other, but there are some differences in capability and performance.”

One of the complexities related to molecular testing is that several viruses can be detected in asymptomatic children.

“We see some of the potential problems in tests that identify DNA or RNA sequences. Although documenting the presence of this or that virus, the tests will not identify what’s along for the ride, ie, an asymptomatic infection, and what is really causing the patient’s problems. Here again, an experienced clinician is invaluable,” Eickhoff said.

Regarding the overall test performance, the risk for a false-positive result in molecular testing is exceedingly low, according to Revell. The real risk is posed with rapid tests, and she said when these tests are done in July for influenza, when there is little or no virus circulating, a positive result is most likely going to be a false-positive. The positive predictive value of these tests changes based on the prevalence of the circulating disease.

“Because the specificity of the molecular test is good, it doesn’t matter what time of year you perform that test,” she said. “The specificity is so high, regardless of the circulating prevalence of the virus. The test is accurate, so if you have a molecular test that says you have flu in July, you probably have flu in July.”

Clinical vs. laboratory diagnosing

Clinical diagnosis is still important, according to Storch, who said that for the foreseeable future, clinical diagnosis will continue to be the primary way viral infections are diagnosed in physicians’ offices and in various other ambulatory centers.

“This is because these [molecular] tests are too expensive, and in most cases, too complicated to do in those settings,” Storch said.

He added that molecular tests will most likely prove useful for children who are admitted to the hospital.

“It will help determine which patients have viral infections; it will help distinguish which viral infection they have. In some cases, especially if it’s influenza, that may dictate therapy with an antiviral drug,” Storch said.

The idea is to “treat the patient, not the chart,” Eickhoff said. “This aphorism was drilled into me as a resident and fellow some years ago, and I – in turn – taught it to a generation of residents and fellows at the University of Colorado. It is reassuring to note that there is still a great deal of truth here – certainly enough to continue to teach it to future generations of physicians… During influenza season, a good clinician can diagnose influenza with an accuracy that exceeds some of the rapid influenza tests that we currently use.”

Laboratory diagnosis also could help with infection control to decide which patients can be roomed with others. Antibiotic usage also could be decreased after a definite diagnosis is established.

Revell also said laboratory diagnostic tests are supplementary to a good clinician: “Certainly, it’s the best way to definitively identify what the etiologic agent is, so when a patient has an upper respiratory infection, the best way to determine what that virus is, is with a molecular test.”

However, when diagnosing for influenza during the winter months, Revell said the tests are not always necessary because if it is influenza season and the patient presents with influenza symptoms, the clinician will be able to recognize the illness.

“But, if there is a question, for example, if you have a transplant patient and maybe things are not as clear, then the definitive diagnostic test is the molecular test,” she said.

Immunocompromised patients

Molecular testing could prove most helpful in immunocompromised patients who develop a variety of infections from viruses, bacteria and fungi.

“Clinical judgment is often not sufficient to make those distinctions,” Storch said. “So here the molecular testing for respiratory viruses can be an important part of the effort to make a diagnosis in those patients.”

There are limitations to these tests, however, Eickhoff said: “We encounter these with increasing frequency as we care for increasing numbers of immunocompromised patients and transplant recipients. Under these circumstances, when clinical expressions of disease may be modified, heightened or suppressed, rapid virologic diagnosis becomes especially valuable.”

According to recent study data, mortality rates in healthy children infected with RSV pneumonia are less than 0.5%, but can reach up to 60% in untreated immunocompromised children. Results from the study, which included 117 RSV-positive immunocompromised pediatric inpatients, found that more than one-third of participants also had a lower respiratory tract infection. Of those with a lower respiratory tract infection, 28% were admitted to the ICU and there was a 5% mortality rate. All of those who died were diagnosed with community-acquired RSV (CA-RSV) lower respiratory tract infection at hospital admission (compared with patients with nosocomially acquired infection).

Pablo Sanchez

Pablo Sanchez

“RSV-related infections in immunocompromised children resulted in poor outcomes, including mortality,” said Sandra Asner, MD, clinical fellow at The Hospital for Sick Children, Toronto. “Mortality was exclusively reported from immunocompromised children with CA-RSV infections. Patients with CA-RSV presented with more advanced clinical evidence of lower respiratory tract disease as compared to those with nosocomial RSV infections.”

Severely immunocompromised patients, such as transplant recipients and many cancer patients, with respiratory virus infections have a high risk for lower respiratory tract infection and death, according to Angela J.P. Campbell, MD, MPH, medical officer of the epidemiology and prevention branch, CDC’s influenza division, and affiliate investigator at the Fred Hutchinson Cancer Research Center.

“When hematopoietic stem cell transplant recipients are infected with RSV, lymphopenia is a risk factor for progression to lower respiratory tract disease, and preemptive therapy with aerosolized ribavirin appears to decrease progression, as well as overall mortality in patients with pneumonia,” Campbell said. “Prompt diagnosis is important for aggressive management.”

However, Campbell added that ribavirin would not be recommended for everyone.

Specifically, with influenza, the earlier treatment begins the better; however, “better late than never,” she said.

“Oseltamivir (Tamiflu, Genentech) dose and duration is uncertain for severe or complicated influenza,” Campbell said. “However, longer duration may be favored in transplant and other immunocompromised patients. More studies are needed to discover the optimal dose, duration and route of antiviral treatment in these patients.”

Parainfluenza virus also poses high risk for severe disease in immunocompromised children.

“[Parainfluenza virus] infections are associated with high morbidity and mortality,” Campbell said. “There are limited treatment options for these infections, and novel therapies need to be developed.”

Under-recognized problem

Respiratory viral infections in the neonatal ICU sometimes go unnoticed until there is an outbreak.

Pablo J. Sanchez, MD, who is a neonatologist and pediatric infectious disease specialist at the University of Texas Southwestern Medical Center, Dallas, said that neonates, including premature babies, are at high risk and are exposed to a number of respiratory viruses from hospital staff, parents and family who visit them.

“There’s a huge knowledge gap in terms of how frequently these viruses do infect babies in the neonatal ICU,” Sanchez said. “We just have not had the technology to accurately identify them earlier and accurately.”

Sanchez said when a cluster of cases is present in the neonatal ICU, an investigation begins.

“The issue in my mind is whether we can identify some of these babies earlier, identify the index case earlier, identify them accurately and therefore be able to intervene and prevent infection in others,” he said.

Sanchez also said the new methodologies of diagnosing respiratory viral infections allow physicians to identify more respiratory viruses that have the potential for spread and cause outbreaks in the neonatal ICU.

“With newer technologies, we are able to identify them and know the cause, as well as potentially be able to stop antibiotics, because it’s not a culture-negative type infection,” he said. “Also, knowing the reason for the infection could potentially prevent some outbreaks by allowing us to take precautions earlier.”

Campbell echoed Revell’s comments about prevention of respiratory virus infections, especially in immunocompromised patients.

“Because existing therapeutic options are inadequate, infection control is critical to prevent transmission,” she said. “Futhermore, for patients with pre-transplant respiratory virus infections, delay of transplantation should be considered when feasible, in an attempt to minimize progression of disease and poor outcomes after transplantation.”

Host vs. pathogen

“In the past few years, there’s been a paradigm shift in how we look at the host immune response,” said M. AsuncionMejias, MD, PhD, principal investigator in the Center for Vaccines and Immunity at The Research Institute and pediatric infectious disease specialist at Nationwide Children’s Hospital, Columbus, Ohio. “We used to think that babies with more severe disease have increased concentrations of different cytokines and markers of inflammation, but actually it doesn’t seem to be the case. It’s the opposite.”

According to Mejias, patients who develop severe RSV disease have a certain degree of “immunoparalysis” and are unable to mount an efficient immune response to fight the infection.

“Intuitively, you would think, that sicker patients have more inflammation, but when infants with RSV [lower respiratory tract infection] develop severe disease and require admission to the ICU, it appears they can’t fight the infection efficiently and actually have less concentrations of proinflammatory cytokines,” she said. “We found an inverse correlation between the cytokine production (TNF-alpha and IL-8) and disease severity.”

Mejias said one of the main problems for physicians is that most of the children who develop severe RSV disease and require hospitalization are previously healthy and do not have any identifiable risk factors clinicians could use to predict who is going to severe illness.

“Actually, we know that there is very little variation in the RSV strain that circulates each season but there is a huge variation in the severity of the disease,” she said. “We’re leaning toward the host playing a leading role in the severity of the disease.” — by Amber Cox


Asner S. Pediatr Infect Dis J. 2013;doi:10.1097/INF.0b013e31829ff4d.
Campbell AP. Respiratory Virus Infections in the Immunocompromised Host: What is Known and What is Needed? Presented at: Pediatric Academic Societies Annual Meeting; May 4-7, 2013; Washington.
CDC. MMWR. 2013;62:141-144.
Mejias A. Markers of Severity and Outcomes in Respiratory Viral infections: Is it the Host, Pathogen, or Both? Presented at: Pediatric Academic Societies Annual Meeting; May 4-7, 2013; Washington.
Sanchez PJ. Respiratory Viruses in the NICU: An Under-Recognized Burden? Presented at: Pediatric Academic Societies Annual Meeting; May 4-7, 2013; Washington.
Storch GA. Advances in the Detection of Respiratory Viruses. Presented at: Pediatric Academic Societies Annual Meeting; May 4-7, 2013; Washington.
Tregoning JS. Clin Microbiol Rev. 2010;23:74-98.

For more information:

Sandra Asner, MD, can be reached at:
Angela J.P. Campbell, MD, MPH, can be reached at:
Andrea T. Cruz, MD, MPH, can be reached at: 6621 Fannin St., Suite A2210, Houston, TX 77030; email:
Theodore C. Eickhoff, MD, can be reached at:
M. Asuncion Mejias, MD, PhD, can be reached at:
Paula Revell, PhD, can be reached at:
Pablo J. Sanchez, MD, can be reached at:
Gregory A. Storch, MD, can be reached at: 660 S. Euclid Ave., Campus Box 8116, St. Louis, MO 63110; email:

Disclosure: The researchers report no relevant financial disclosures.

Should molecular testing be used for all children admitted to the hospital with respiratory viral infections?


Sensitivity and specificity of the tests justify the cost.

In recent years, molecular tests for respiratory viruses have become increasingly available in kit format in many centers. These FDA-approved tests allow rapid detection of multiple respiratory viruses (and some non-viruses, such as Mycoplasma. pneumoniae, Bordetella pertussis, Chlamydia pneumoniae) with very high sensitivity and specificity and without the requirement for extensive laboratory personnel time or expertise. Now that they are available, it remains to be determined how such tests should best be utilized. Should they be considered for all children hospitalized with respiratory infections?

Information gleaned from such testing has a number of potential benefits:

1. Decrease or prevent unnecessary antibiotic use.

2. Shorten hospital length of stay.

3. Promote appropriate antiviral therapy (eg, neuraminidase inhibitor for influenza; possible antiviral treatment of respiratory syncytial virus, adenovirus or parainfluenza infections in the immunocompromised host).

4. Guide appropriate infection control measures and help minimize nosocomial transmission of infection.

5. Educate families and staff about specific viral agents and their natural history.

6. Improve the understanding of infections with specific agents and the role of viral coinfections.

Admittedly, these tests are expensive and further studies are needed to confirm these potential benefits in order to offset the cost of the tests and determine settings where they can be most helpful. Will patients do better with the availability of such testing? Will hospitalizations be shortened or prevented? Will the results of molecular viral tests lead to a decrease in unnecessary antibiotic usage or other ancillary tests? How do we interpret specimens with multiple viral agents identified?

There are data that suggest significant benefits can be achieved with use of such tests. Enteroviral testing by PCR has been shown to be cost-effective (Ramers C. JAMA. 2000;283:2680-2685). Colvin and colleagues (Pediatrics. 2012:130:e1455-1462) have demonstrated the ability to identify viral pathogens in young children with fever without an apparent source in an emergency room setting, which could potentially prevent unnecessary antibiotic use and possibly avoid hospitalizations in some instances. Finally, we have new data to be presented in the fall (Makvana S. ID Week; October 2013, San Francisco) on a quality performance initiative to successfully decrease inappropriate antibiotic administration in RSV-positive infants relying in large part on specific viral identification.

Further large-scale studies will help identify whether these benefits will be meaningful for all or most children hospitalized with possible respiratory viral infection, but certainly in a number of instances they promise to provide clinically important information that can benefit children and justify their cost.

Leonard R. Krilov, MD, is chief of pediatric infectious diseases, Winthrop University Hospital, Mineola, N.Y. Disclosure: Krilov reports no relevant financial disclosures.



Infection control can be more effective than expensive testing.

The viruses known to cause infection of the respiratory tract of children continue to grow and include respiratory syncytial virus; rhinoviruses; parainfluenza viruses; influenza viruses; human metapneumovirus; coronaviruses; enteroviruses; adenoviruses; and human bocaviruses. With the availability of molecular amplification techniques, nasopharyngeal aspirates can be analyzed quickly for the presence of one or more viral pathogens (as well as other pathogens, such as M. pneumoniae, B. pertussis, C. pneumoniae). The important question becomes, "Is the expense of such assays justified by the information gained?"

The answer to this question may depend on why the assay is being performed. In those settings where clinical research is being performed or for epidemiologic monitoring purposes, understanding the specific etiology is important. In acutely ill children who may be immune compromised or who have underlying conditions, knowledge of the etiology may be helpful for treatment decisions (such as consideration of cidofovir (Vistide, Gilead) for adenovirus pneumonia).

Viral testing of all children with respiratory tract infection is unlikely to be cost-effective. For hospitalized or outpatient children, rapid antigen detection assays for influenza may be useful in deciding whom to treat. The reliability of antigen screening tests are highly variable depending on factors such as the prevalence of the virus in the community, the duration of the child’s illness and the adequacy of the specimen obtained. Nucleic acid detection assays have the advantage of greater sensitivity than viral culture or antigen detection. But the ability of nuclear amplification techniques to detect respiratory virus(es) in a child who has no symptoms brings into question the clinical significance of a positive result. Perhaps a positive result from either a non-symptomatic or a symptomatic child indicates colonization or residual virus from the last illness rather than active infection.

Infection control practices are critical to prevent nosocomial spread of respiratory viruses. Most respiratory viruses spread among hospitalized children as large particle aerosols that travel less than 3 feet. Optimal infection control measures for any symptomatic child are likely to be more important than expensive molecular testing. This may change when specific antiviral therapy becomes available for respiratory viruses.

H. Cody Meissner, MD, is head of the division of pediatric diseases at the Floating Hospital for Children, Tufts Medical Center, Boston. He is also a member of the Infectious Diseases in Children Editorial Board. Disclosure: Meissner reports no relevant financial disclosures.