点击显示 收起
1Institute of Virology, University of Vienna, and 2St. Anna Children's Hospital, Vienna, Austria
Received 26 July 2002; revised 3 October 2002; electronically published 6 January 2003.
Serum samples from 68 immunocompetent infants (mean age, 12.6 months) with an acute adenovirus infection of the respiratory tract (39 experiencing their first adenovirus infection) were tested for the presence of adenovirus DNA, to investigate whether viral dissemination via the blood is usually present in the immunocompetent patient. Using a nested polymerase chain reaction assay, adenovirus DNA could be detected in acute-phase serum samples from 28 (41%) children. Adenovirus DNA was never found in follow-up serum samples, indicating a short period (1 week) of viral dissemination. In children experiencing their first adenovirus infection, viral DNA could be detected in 72% of the acute-phase serum samples collected within the first week after onset of symptoms. Adenovirus DNA could also be detected in 25% of the acute-phase serum samples from patients with reinfection.
Presented in part: winter meeting of the European Society for Clinical Virology, Rotterdam, The Netherlands, 79 January 1999 (abstract 68).
The study was approved by the ethics committee of the St. Anna Children's Hospital, and informed consent was obtained from the parents of all participating children.
Reprints or correspondence: Dr. Stephan W. Aberle, Institute of Virology, University of Vienna, Kinderspitalgasse 15, A-1095 Vienna, Austria .
Adenovirus infections of the respiratory tract are exceedingly common. Antibodies to 1 of the serotypes can be detected in 70%80% of children at age 5 years. The clinical course of adenovirus infections in immunocompetent children is usually benign, and an upper respiratory tract illness is most commonly observed [1]. Involvement of the lower respiratory tract may result from the progression of local infection or may be the effect of viremia. There is some evidence for the presence of viremia in immunocompetent patients from the appearance of maculopapular, morbilliform, and petechial exanthemas in the course of adenovirus infections [2]. In addition, the virus has been recovered from peripheral blood and multiple organs from children with disseminated adenovirus disease [35]. Although there are reports that the detection of viral DNA in peripheral blood provides an excellent marker for viral dissemination in the course of adenovirus infection in immunosuppressed patients [6, 7], data on the presence and duration of adenovirus viremia in immunocompetent patients are not available. We therefore decided to use a nested polymerase chain reaction (PCR) for the sensitive detection of adenovirus DNA in serum samples, to investigate whether dissemination of adenoviruses via the blood can be detected in immunocompetent children with an acute adenovirus infection of the respiratory tract.
Materials and methods. Of 3020 children hospitalized with an acute respiratory tract infection, an adenovirus infection was confirmed in 143 (4.7%; mean age, 13.6 months). Serum samples stored at -20°C were available for testing from 68 children (mean age, 12.6 months; range, 148 months). According to the findings of physical examination and chest radiographs, 21 of these patients had upper respiratory tract infections (URTIs), 47 had lower respiratory tract infections (LRTIs), 8 had bronchitis, 26 had obstructive bronchitis, and 13 had pneumonia. All of them had a body temperature >38°C at admission.
Adenovirus infection was confirmed by the presence of adenovirus in nasopharyngeal aspirates (NPAs), using virus isolation in tissue culture and antigen detection by means of ELISA (Ag-ELISA). These NPAs were all negative for other respiratory viruses, such as influenza viruses A and B; parainfluenza 1, 2, and 3; rhinoviruses; respiratory syncytial virus (RSV); and enteroviruses, when tested by virus isolation and Ag-ELISA techniques. Acute-phase serum samples were drawn at the same time as NPAs 125 days (median, 5 days) after the onset of symptoms, as determined by chart review. Twenty-seven convalescent-phase serum samples (23 second and 4 third serum samples taken 662 days [median, 11 days] and 1869 days [median, 42 days] after the onset of symptoms, respectively) were available for testing from 23 patients. The presence of adenovirus-specific IgG and IgM antibodies was determined by means of a commercially available antibody ELISA (Sorin Biomedica Diagnostics), used according to the manufacturer's instructions.
The rhinovirus-sensitive HeLa cell "Ohio" strain [8], kindly provided by Dr. David A. J. Tyrrell (Clinical Research Centre, Common Cold Unit, Salisbury, Wiltshire, UK), was used for virus isolation, as described elsewhere [9]. Adenoviruses; RSV; parainfluenza virus 1, 2, and 3; and influenza virus A and B were further identified by indirect immunofluorescent antibody assay. Enteroviruses and rhinoviruses were classified using the acid lability test. The ELISA for the detection of adenovirus; RSV; parainfluenza virus 1, 2, and 3; and influenza virus A and B antigens in NPAs was performed as described elsewhere [10, 11].
DNA extraction from 200 of L serum was performed with QIAamp DNA Blood Mini Kit (Qiagen), according to the manufacturer's instructions. DNA was resuspended in 200 L of distilled water.
The nested PCR for the detection of all adenovirus serotypes was carried out using primers located within the highly conserved hexon gene, as described by Allard and Wadell [12]. The first (nested) step amplification was carried out in a 50-L reaction mixture that contained 10 L of template DNA (2 L of amplicon), 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2 mM MgCl2, each dNTP at 200 M, outer (inner) primers at 25 pmol, and 1 U Taq-Gold DNA polymerase (Perkin-Elmer/Cetus). The thermocycler profile was 95°C for 10 min, 35 cycles of 95°C, 60°C, and 72°C each for 30 s, and a final extension at 72°C for 5 min. The presence of 143-bplong amplicons visualized on a 3% NuSieve gel (FMC Bioproducts) indicated a specific positive result. To monitor for contamination, numerous negative water controls were included in every run. Every positive PCR result was confirmed by testing a second portion of the original serum sample. False-negative PCR results due to unspecific inhibition of amplification were excluded by spiking 8 L of the extracted DNA-negative samples with 2 L of template DNA that contained 10 copies of adenovirus standard. All spiked samples were found to be positive for adenovirus DNA. The detection limit of the adenovirus PCR was 100500 copies/mL, obtained by testing dilutions (in distilled water) of the commercially available, particle-counted, adenovirus subtype 5 standard (Advanced Biotechnologies). Comparison of the 2 groups was carried out using the Mann-Whitney U test, and proportions were compared using the 2 and Fisher's exact tests, as appropriate.
Results. In 28 (41%) of 68 children with an acute respiratory adenovirus infection, adenovirus DNA could be detected in the acute-phase serum sample when a nested PCR was used. Data on the number of days after onset of symptoms, clinical diagnosis, adenovirus-specific antibody status, and age of the children with and without adenovirus DNA detectable in acute-phase serum samples are summarized in .
fig.ommitted
Table 1. Characteristics of adenovirus DNApositive children with acute adenovirus infections.
Serum samples from children positive for adenovirus DNA were drawn an average of 2 days earlier (median, 4 days; range, 119 days) than were serum samples from children with a negative PCR result (median, 6 days; range, 125 days) (P = .046, Mann-Whitney U test). Twenty six (93%) of the adenovirus DNApositive samples were drawn within the first week after the onset of illness, compared with 2 DNA-positive samples drawn later in the course of the disease (P = .002, Fisher's exact test). No correlation could be observed between the clinical diagnosis of URTI or LRTI and the detectability of adenovirus DNA in the acute-phase serum samples.
The majority, 75% (21/28), of the adenovirus DNApositive children had no detectable adenovirus-specific IgG antibodies in their acute-phase serum samplesthat is, an adenovirus antibody level <30 arbitrary units (AU), indicating primary infection. When all the children with a primary adenovirus infection were taken into consideration, 54% (21/39) had detectable adenovirus DNA in their acute-phase serum samples. Within the first week after onset of primary adenovirus infection, adenovirus DNA could be detected in the serum of 72% (21/29) of the children. The majority of primary infections were found in infants 712 months old, an age group in which maternal antibodies have mostly dropped below detectable levels and primary infection usually takes place.
Adenovirus DNA was detectable in the acute-phase serum of 7 children in the presence of adenovirus-specific IgG antibodies. The high levels of adenovirus-specific IgG antibodies (mean, 860 AU) and a mean age of 19 months in 5 of these patients indicate that adenovirus DNA is present in the serum of patients undergoing reinfection. The remaining 2 children, who were 7 and 9 months old, had antibody levels of 35 and 155 AU, respectively. In one of them, the serum sample was drawn 8 days after the onset of symptoms. In the other infant, hospitalization because of an acute URTI 3 days before the collection of the first serum sample indicates more recent adenovirus infection than the 19 days reported .
fig.ommitted
Figure 1. Presence of adenovirus DNA in first, acute-phase and follow-up serum samples in relation to adenovirus-specific antibodies and the no. of days after onset of symptoms. Black bars, detectable adenovirus DNA; gray bars, nondetectable adenovirus DNA; solid bars, no adenovirus IgG antibodies (<30 arbitrary units [AU]); hatched bars, presence of adenovirus IgG antibodies (30 AUs). The cross represents positive results of testing for adenovirus IgM antibodies.
A total of 27 follow-up serum samples from 23 infants (10 with detectable DNA in the first serum sample and 13 without) were available for testing . Adenovirus DNA was never found in follow-up serum samples.
Seroconversion was observed between days 10 and 69 (median, day 15) after onset of symptoms in 9 of 16 children experiencing their first adenovirus infection. In the remaining 7 infants, seroconversion could not be demonstrated in the follow-up samples available, which had been drawn 612 days (median, 10 days) after the onset of symptoms. Adenovirus-specific IgM antibodies were only found in 3 infants and with levels near the cutoff value . All of these children were adenovirus PCR negative.
Discussion. Our results demonstrate that adenovirus DNA can be detected in the majority (72%) of serum samples collected within the first week after the onset of symptoms from immunocompetent infants experiencing their first episode of acute respiratory adenovirus infection. Adenovirus DNA was never detectable in follow-up serum samples, which indicates a shorter duration (1 week) of the viremic phase in immunocompetent children, compared with that observed in immunosuppressed patients [6, 7]. The following explanations are conceivable for the failure to detect adenovirus DNA in all the acute-phase serum samples from children with a primary infection: individual variability in the control of viral replication, later sampling and incorrect reporting on the number of days after onset of symptoms, and, finally, a virus load below the detection limit. Successful containment of viral replication depends on the generation of antigen-specific T cells and may, therefore, vary among individuals [13]. In addition, acute-phase serum samples from DNA-negative infants were usually drawn later, after the onset of symptoms, and viral clearance by cytotoxic T cells may already have been completed. It may also be that the number of days after the onset of symptoms reported is not always related to the viral pathogen detected, especially in infants of that age, who usually experience multiple consecutive respiratory tract infections within a relatively short period of time [14].
During the course of primary infection, adenovirus-specific IgG antibodies began to become detectable by the end of the second week (approximately days 1214) after the onset of symptoms. Virus-specific IgM antibodies were only detectable in 3 infants with confirmed primary infection; therefore, detection of such antibodies cannot be recommended as a reliable diagnostic tool.
Viral dissemination also seems to occur during the course of reinfections, as indicated by the detection of viral DNA in the presence of high levels of virus-specific IgG antibodies in the acute-phase serum samples of 5 children with a median age of 19 months. Nevertheless, this number is too small for conclusions to be drawn about the frequency with which detectable levels of DNA are found in the serum samples from patients experiencing reinfections.
Two infants with detectable DNA and IgG antibodies differed from those 5 classified as having reinfection by their younger age and by their comparatively lower antibody levels. This argues for an adenovirus primary infection in the presence of decreasing levels of maternal antibodies rather, than for viral dissemination in the course of reinfection.
Adenovirus DNA could not be detected in the majority of the acute-phase serum samples with elevated levels of virus-specific IgG antibodies. This may be due to a significantly lower virus load and/or a shorter viremic phase in the course of reinfection. Because there is evidence that adenoviruses can be excreted in nasopharyngeal secretions for a prolonged period of time [15], it is also conceivable that in, some of the infants, the adenovirus detected in nasopharyngeal secretions was not the cause of the recent acute respiratory tract infection.
In conclusion, the results of our study clearly demonstrate that viral dissemination is a common event in immunocompetent infants passing through their first adenovirus infection but can also be detected in 25% of children experiencing reinfection. Knowledge of the frequency and duration of viral dissemination in immunocompetent infants provides a basis for the interpretation of data obtained in immunosuppressed patients.
Acknowledgments
We thank Michaela Binder, Barbara Dalmatiner, Sylvia Malik, and Ursula Sinzinger for their excellent technical assistance. We are also grateful to Annemarie Witzelsberger for the collection of specimens.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.