The predominant mode of human influenza transmission is droplet-borne and small particle aerosols. Viral shedding profiles of influenza infections might provide helpful information for understanding virulence, cell tropism, transmission dynamics and designing management policies.
MATERIALS AND METHODS
Children younger than 17 years admitted to Children’s Memorial Hermann Hospital, Houston, TX, between January 2009 and April 2011 with laboratory-confirmed influenza infection were eligible for enrollment. Informed written consent was obtained from the parent or legal guardian. Data were collected using a standard case report form. The study was approved by the Institutional Review Board for the University of Texas Health Science Center at Houston and Baylor College of Medicine.
We attempted to collect 3 nasopharyngeal wash (NPW) specimens, 3 stool specimens and 1 blood sample from each patient. Blood specimens were collected in ethylenediaminetetraacetic acid tubes with plasma and red blood cell stored separately at −70°C after centrifugation.
NPW specimens were collected using a standard protocol.1 Stool specimens were immediately diluted (10% solution) with viral transport medium and subsequently clarified by low speed centrifugation and filtration (0.45 micron filter) and stored at −80°C. All samples were tested in duplicate using quantitative polymerase chain reaction (qPCR) for seasonal influenza A and B and novel H1N1 (nH1N1).1 Specimens were considered positive if its Ct value occurred before 40.0 cycles. Blood and stool samples were spiked with known concentration of hMPV to serve as an internal control. All samples were placed on rhesus monkey kidney cells for virus isolation.2
Twenty subjects were enrolled: median 1.6 years (range 0.1–15). Eleven were male, and 8 were Hispanic. Twelve had underlying pulmonary conditions. Of 18 vaccine eligible children, vaccine history was verified in 17. Only 3 had received the recommended influenza vaccination(s) for their age group. Common presenting symptoms were respiratory distress (14, [70%]), gastrointestinal (GI) symptoms (12, [60%]) and fever (12, [60%)]). Median hospital duration was 3.5 days (range 1–22). Four required intubation, and 6 were admitted to the pediatric intensive care unit. Fourteen were treated with oseltamivir with most (9, [64.2%]) receiving it within 48 hours.
Seventy-four specimens were collected: 35 NPW, 29 stool and 10 blood (Table 1). The mean period between clinical onset and initial NPW collection was 4.8 days, 4.5 days for stool and 5.3 days for blood. Sixty-four specimens (35 NPW and 29 stools) were tested by cell culture. Influenza virus was isolated from 10 (28.6%) NPW specimens: 7 were nH1N1 and 3 influenza B. Seventy-four samples were tested by qPCR and 41 (55.4%) were positive: 32 NPW (21 nH1N1 and 11 seasonal [7 influenza A related to seasons 2008 to 2009; 4 influenza B related to season 2010 to 2011]), 1 blood sample (red blood cell, nH1N1) and 8 stool (all nH1N1) (Table 1).
Ten of 27 qPCR NPW samples with Ct value of ≤35 were culture positive for influenza compared with 0 of 5 with Ct value >35. Seven of 8 qPCR stool samples had Ct values >35, and none was culture positive for influenza. The median NPW Ct value was 31.6 for seasonal influenza (A and B), 27.5 for nH1N1 and the median stool Ct value was 36.8. The median Ct values for the initial NPW and stool specimen did not differ by symptom duration.
Of 12 children with GI symptoms, only 2 had nH1N1 detected by qPCR in their stool. Three of 5 children with qPCR-positive stools did not have GI symptoms. All 5 children with qPCR-positive stools had qPCR-positive NPW samples with correlating influenza types. In addition, 3 of the 5 children had >1 positive stool sample. There were no significant differences between children with and without influenza detected in stool with respect to age, underlying diseases, hospital duration, intensive care unit stay, ventilator use or NPW Ct value (33.04 vs. 28.49, P value 0.17). Of note, children without evidence of influenza RNA in the stool had over a 23-fold increase in influenza RNA in their NPW samples (correlating NPW sample if stool positive, first positive NPW sample if stool negative) compared with children who had influenza RNA in stool. One immunocompetent child had qPCR-positive blood. This patient also had a correlating qPCR-positive NPW sample.
We describe the clinical features and viral shedding patterns in respiratory and nonrespiratory sites using qPCR and culture. Over half of the children presented with GI symptoms. GI symptoms, such as anorexia, nausea, vomiting and diarrhea, are known to be common in children with influenza infection. In a review of 84 children hospitalized with influenza A virus (1997 to 2002), vomiting (35%) and diarrhea (18.4%) were often found in younger children.3 Reviews of children hospitalized with nH1N1 have reported GI symptoms occurring up to 52.4%.4 As was the case with our study, the majority of these children were young.
GI symptoms may be a hallmark of severe influenza disease in children. Of 153 children who died from influenza during the 2003 to 2004 season in the United States, 46% presented with GI symptoms.5 Many of these children had underlying conditions known to increase the risk of influenza complications; however, 47% were healthy. None of the children in our study died; however, 6 (30%) were admitted to the pediatric intensive care unit with 2 (33.3%) having GI symptoms. Like that in Bhat report, only 3 of the 18 vaccine eligible children in our study had received age-appropriate vaccination.
We detected influenza in both nonrespiratory sites studied, stool and blood. Detecting influenza RNA in stool has previously been reported6,7 with stool detection rates of up to 24.6%6 and 47%.7 Similar to our results, viral RNA positivity had little correlation with GI symptoms or outcome.6,7 Unlike prior reports, fecal viral concentration in our study did not correlated with symptom duration.7 This could have been related to our small sample size. Time from collection to storage would have been less likely to impact results as stool specimens were process more quickly for RNA-negative (26.0 hours [range 0.2–69]) versus RNA-positive (41.6 [2–72]) samples. We were not able to isolated live influenza virus from stool. This may have been related to the low inoculum or dilution effect inherent in stool samples. This is supported by the high Ct values detected in 7 of 8 qPCR-positive stool samples.
To our knowledge, we report the first case of nH1N1 viremia in an immunocompentent child. Viremia with seasonal influenza has been rarely reported. In contrast, viremia with nH1N1 has been reported in the blood of patients with severe infection8 and among immunocompromised patients9 using serum or plasma. Animal models have suggested red blood cell, such as we used in our study, to be a more successful target for polymerase chain reaction.10 Our enrollment period (January 2009 to April 2011) involved 3 influenza seasons; however, only children with nH1N1 had viral RNA detected in nonrespiratory sites. Thus, our clinical and laboratory findings may be more reflective of the nH1N1 pandemic virus, highlighting perhaps a difference in cell tropism between seasonal and nH1N1 influenza.
In conclusion, we detected viral RNA in respiratory and nonrespiratory sites among immunocompetent children. Influenza RNA in stool was not associated with the presence of GI symptoms or more severe disease. Cultivable influenza viruses were not detected in stool; however, the presence of viral RNA raises infection control concerns. The finding of viremia in an immunocompetent child adds to the potential for systemic spread to nonrespiratory sites during influenza infection in children and adverse outcome.
1. Mansbach JM, Piedra PA, Stephen SJ, et al. Prospective, multicenter study of viral etiology and hospital length of stay in children with severe bronchiolitis. Arch Pediatr Adolesc Med. 2012;166:700–706
2. Glezen WP, Gaglani MJ, Kozinetz CA, et al. Direct and indirect effectiveness of the influenza vaccination delivered to children at school preceding an epidemic caused by 3 new influenza virus variants. J Infect Dis. 2010;202:1626–1633
3. Wang YH, Huang YC, Chang LY, et al. Clinical characteristics of children with influenza A virus infection requiring hospitalization. J Microbiol Immunol Infect. 2003;36:111–116
4. Tran D, Vaudry W, Moore DL, et al. Comparison of children hospitalized with seasonal versus pandemic influenza A, 2004–2009. Pediatrics
5. Bhat N, Wright JG, Broder KR, et al. Influenza Special Investigations Team. Influenza-associated deaths among children in the United States, 2003–2004. N Engl J Med. 2005;353:2559–2567
6. Yoo SJ, Moon SJ, Kuak EY, et al. Frequent detection of pandemic (H1N1) 2009 virus in stools of hospitalized patients. J Clin Microbiol. 2010;48:2314–2315
7. Chan MC, Lee N, Chan PK, et al. Seasonal influenza A virus in feces of hospitalized adults. Emerg Infect Dis. 2011;17:2038–2042
8. Tse H, To KK, Wen X, et al. Clinical and virological factors associated with viremia
in pandemic influenza A/H1N1 /2009 virus infection. PLoS One. 2011;6:e22534
9. Choi SM, Xie H, Campbell AP, et al. Influenza viral RNA detection in blood as a marker to predict disease severity in hemtopoietic cell transplant recipients. J Infect Dis. 2012;206:1872–1877
10. Likos AM, Kelvin DJ, Cameron CM, et al.National Heart, Lung, Blood Institute Retrovirus Epidemiology Donor Study-II (REDS-II). Influenza viremia
and the potential for blood-borne transmission. Transfusion. 2007;47:1080–1088