262. Diagnostic Utility of Human Parechovirus (HPeV) and Enterovirus (EV) RT-PCR Testing on non-CSF Specimens in Infants Less than 90 Days of Age
Session: Poster Abstract Session: Diagnostic Microbiology; Novel Molecular Methods
Thursday, October 3, 2013
Room: The Moscone Center: Poster Hall C
Background: Human Parechoviruses (HPeVs) are recognized causes of infant sepsis-like illness and CNS infection. No studies have examined testing utility of non-CSF specimens for HPeV. Our aim was to evaluate non-CSF specimens in predicting HPeV or EV CNS infection in infants with sepsis-like illness. 

Methods: Hospitalized infants < 90 days old with CSF WBC counts <1000 and negative CSF gram stain were enrolled between 1/3/2011–12/31/2012. Stool, throat and nasal specimens were prospectively obtained. Scavenged CSF, blood and urine were collected when available. EasyMag® or Qiacube® was used to extract total nucleic acids, which were then tested by two-step real-time EV/HPeV RT-PCR. Diagnostic utility of non-CSF specimens to predict HPeV-CNS infection was determined by analyzing EV/HPeV results from paired CSF and non-CSF specimens in the same infants. 

Results: From 461 infants, 2001 specimens were obtained (432 CSF, 341 blood, 224 urine, 407 throat, 229 nasal, 368 stool). HPeV was detected in 164 specimens (46 subjects) and EV in 115 specimens (47 subjects). HPeV and EV were detected in 40 and 35 CSF samples, respectively. HPeV detection sensitivity in non-CSF specimens compared to CSF detection: blood (96%), stool (82%), throat (71%), nasal (63%) and urine (43%). HPeV detection specificity in all non-CSF specimens was 100%. EV detection sensitivity in non-CSF specimens: stool (94%), blood (64%), throat (41%), urine (6%) and nasal (5%). EV detection specificity for all non-CSF specimens was >95%. HPeV vs. EV blood and respiratory specimen testing showed HPeV blood testing outperformed EV blood testing in correctly predicting CNS disease presence/absence (p = 0.037), as did HPeV respiratory testing vs. EV respiratory testing (p = 0.015).        

Conclusion: HPeV can be detected from multiple anatomic sites outside CSF in HPeV-infected infants. Excluding stool, HPeV is detected at a higher frequency in non-CSF specimens than EV. Blood appears a useful non-CSF specimen for HPeV (96%) and EV (64%) detection in infants with sepsis-like illness. Stool appears suitable for both HPeV and EV detection, while respiratory specimens (nasal/ throat) are more reliable for infant HPeV detection. PCR testing of non-CSF specimens can be useful adjuncts in diagnosis of HPeV CNS disease in infants presenting with sepsis-like illness.

J. Michael Klatte, M.D.1, Christopher Harrison, M.D.2, Brian Pate, M.D.1, Mary Anne Jackson, MD, FIDSA1 and Rangaraj Selvarangan, PhD1, (1)Children's Mercy Hospitals and Clinics, Kansas City, MO, (2)Pediatrics, Children's Mercy Hospitals and Clinics, Kansas City, MO


J. M. Klatte, None

C. Harrison, None

B. Pate, None

M. A. Jackson, None

R. Selvarangan, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.