点击显示 收起
Departments of Pediatrics and Preventive Medicine, University of Tennessee School of Medicine, University of Tennessee Graduate School of Health Sciences, Le Bonheur Children's Medical Center, The Children's Foundation Research Center
St. Jude Children's Research Hospital, Memphis
Background.
Elucidating the relationship between viral load and respiratory syncytial virus (RSV) disease severity is critical to understanding pathogenesis and predicting the utility of antivirals.
Methods.
Previously healthy, naturally RSV-infected infants <24 months old not treated with ribavirin, passive antibody, or corticosteroids were prospectively studied (n = 141). Viral loads were measured by fresh quantitative culture from nasal washes collected at a single time point shortly after hospitalization.
Results.
The subjects' mean age was 112.1 days, and the mean estimated gestational age at birth was 38.38 weeks. RSV load decreased with longer durations of symptoms before specimen collection (P = .01). Male subjects had higher RSV loads than female subjects (P = .036). Significant independent predictors of longer hospitalization were congenital anomaly (P < .0001), lower weight on admission (P = .028), and higher nasal RSV load (P = .008). A 1-log higher RSV load predicted a 0.8-day longer hospitalization. Lower weight and higher RSV load were also independently associated with respiratory failure (P < .0005 and P = .0049, respectively) and requirement for intensive care (P = .0007 and P = .0048, respectively).
Conclusions.
In previously healthy infants, higher RSV loads measured at capturable time points after symptom onset predict clinically relevant measures of increased disease severity.
Respiratory syncytial virus (RSV) is the most common cause of lower respiratory tract infection in the developed and the developing world. In the United States, 3% of healthy full-term infants are hospitalized for RSV infection [1, 2], making it the single most common cause of hospitalization in this age group [2]. Although infants who have chronic lung disease, congenital heart disease, or immunodeficiencies or were born prematurely are at increased risk for severe RSV disease, most infants hospitalized with RSV infection are previously healthy [1, 3, 4]. Within this previously healthy population, there is a wide range of RSV disease severity, and the factors predisposing to severe disease have been scrutinized.
The pathogenesis of RSV infection is thought to be largely due to the host immune response [57], for which disease enhancement following the formalin-inactivated RSV vaccine is a poignant general example [8]. Cell-mediated immune responses [9] and, more specifically, an inappropriate shift toward a Th2 and away from a Th1 response to RSV infection has been observed to produce more severe disease in animal models [10] and may also be relevant in humans [11]. Other mechanisms that involve a pathogenic immune response include immune complex formation [12] and bystander killing effects of cytotoxic T cells [13]. However, less attention has been given to the role that viral replication plays in disease severity.
Quantitative differences in the replication of RSV may alter the magnitude of later downstream immunopathological events. In addition, the relationship between viral replication and RSV disease severity is important in the determination of the potential utility of antiviral treatment strategies. We therefore sought to study the RSV load in naturally infected infants and to determine its relationship to clinically relevant markers of disease severity.
SUBJECTS, MATERIALS, AND METHODS
Subjects.
We conducted a prospective observational study of infants infected with RSV during 4 successive winter seasons (19972001). Clinical data were obtained through interviews of each subject's caregiver. Subjects were included if they were <2 years old and had RSV detected 48 h before enrollment. Subjects were excluded if they had a diagnosed immunodeficiency, had received systemic corticosteroids within the preceding month, had diagnosed chronic lung disease of prematurity (bronchopulmonary dysplasia), had hemodynamically significant congenital heart disease, or had positive blood cultures. Disease severity was determined by the duration of hospitalization and whether subjects required intensive care or had respiratory failure during their hospitalization. To determine the timing of symptoms, the caregiver was asked to provide the date of onset of each of the individual symptoms: fever, runny nose, cough, and difficulty breathing. Of these symptoms, the one with the earliest onset was chosen for use in the calculation of the duration of symptoms prior to enrollment. The study was conducted with the approval of the University of Tennessee Institutional Review Board, and appropriate informed consent was obtained. The human experimentation guidelines of the US Department of Health and Human Services were followed in the conduct of this research.
Definitions.
Respiratory failure was defined as the requirement for mechanical ventilation. Subjects who were on mechanical ventilation solely for apnea with a fraction of inspired oxygen 0.30 were not counted as having respiratory failure. Tobacco exposure was defined as being present if any person living in the subject's household smoked tobacco, regardless of whether smoking was limited to outside the subject's house. Breast-feeding was defined as being present if the subject was fed by breast at all. Limited breast-feeding that occurred only during birth hospitalization was not included. Day care attendance was defined as any current attendance in a group care setting that had >5 children.
Collection of nasal wash specimens.
Trained study personnel collected nasal aspirates by use of a standard quantified technique. Briefly, subjects were placed supine. Nonbacteriostatic saline was instilled into one naris, an 8 French catheter was immediately inserted to a depth of 5 cm, and suction aspiration was performed during removal of the catheter. The procedure was repeated twice on each naris (total instilled saline volume, 3 mL). Cold RSV stabilization medium (3 mL) (EMEM, 7.5% sucrose, 25 mmol/L HEPES, 1% L-glutamine, and 1% penicillin/streptomycin) was immediately aspirated through the suction catheter and was mixed with the subject's secretions. Stabilized specimens were transported quickly on wet ice to the laboratory for processing.
Virologic methods.
Preliminary identification of RSV-infected subjects was made by use of either direct fluorescent antibody (DFA) (Bartels; Trinity Biotech) or a rapid antigen detection assay (Directigen; Becton Dickinson) [14, 15]. Less than 5% of these identifications were performed with Directigen. A HEp-2 cell plaque assay in 12-well plates with 10-fold dilutions in triplicate was used. Plates were fixed and stained with hematoxylin-eosin 5 days after inoculation. Plaque assays were performed on fresh specimens within 3 h of collection. HEp-2 cells passaged <18 times were used for all experiments. To assure quantitative reliability, RSV quantitative standards were run in parallel with each plaque assay [16]. The RSV quantitative standards were individual-use aliquots from the supernatant of RSV A Long (ATTC VR-26) grown in HEp-2 cells and frozen at -80°C in 25% sucrose. Each individual aliquot was thawed and diluted 1000-fold in EMEM (with 1% L-glutamine, 25 mmol/L HEPES, 1% penicillin/streptomycin, and 5% fetal bovine serum) before it was used as the parallel quantitative standard. A subject was defined as being infected with RSV if both the antigen detection assay was positive and the subject's nasal wash plaque assay produced RSV cytopathic effect.
Statistical analysis.
Statistical analysis was performed with standard methods using SAS software (version 10.5; SAS Institute). Univariate, descriptive statistics and graphs were prepared using Prism software (version 4.0; GraphPad). All significance tests were 2-tailed ( < 0.05). For normally distributed data, Student's t test was used. For data not normally distributed, the Mann-Whitney U test was used. When 2 dichotomous variables were being related, Fisher's exact test was used. For the multiple regression and logistic regression analyses, a set of a priori predictive models were constructed to predict the individual disease severity markers of (1) duration of hospitalization, (2) requirement for intensive care, and (3) respiratory failure. For each marker of disease severity, the model that had the lowest corrected Akaike's information criteria value was chosen as the final model. The model for the duration of hospitalization included the following variables: age, sex, race/ethnicity, prematurity (<38 weeks), breast-feeding, congenital anomalies, tobacco exposure, day care attendance, weight on admission, the number of persons living in the subject's home, the duration of symptoms prior to collection of the nasal aspirate specimen, the duration of symptoms before admission to the hospital, and the RSV load. The variables included in the model to best predict a requirement for intensive care were all of these except congenital anomalies, because only 1 subject with congenital anomalies required intensive care. The variables included in the model to predict respiratory failure were the same as those used to predict the requirement for intensive care, except that day care attendance was not included in the model, because only 1 day care attendee had respiratory failure.
RESULTS
The prospective observational study enrolled 188 RSV-infected infants during 4 successive winter seasons (19972001). Thirty-seven infants were excluded due to unrecognized single or multiple occurrences of the following: 13 had received corticosteroids, 22 had received RSV-specific passive antibody, 1 had congenital heart disease, and 11 had chronic lung disease of prematurity. Of the remaining subjects, 10 of 151 nasal wash plaque assays failed to produce RSV cytopathic effect. Thus, 141 subjects met the entry criteria and constituted the analyzed data set.
Clinical characteristics of the study population.
Descriptors of the 141 study subjects and their disease severity risk factors are shown in table 1. The study population consisted of predominantly full-term infants with a mean estimated gestational age at birth of >38 weeks; only 9.22% of the subjects had an estimated gestational age of <35 weeks. The weight at birth (mean, 3037 g) also confirmed that the majority of subjects were full-term, healthy infants. The youngest subject was 10 days old, and 25% of the subjects were <41 days old. Reflecting the racial/ethnic distribution of the population served by the enrolling medical center, 36 of the subjects were white, 1 subject was of mixed race, and the remaining subjects were African American. Only 3 subjects had congenital anomalies. The RSV disease characteristics of the study subjects (table 2) were similar to those seen in other infants hospitalized with RSV [4, 17]. Of the 141 subjects, 19.86% required intensive care. No deaths occurred during the study. No subject received passive RSV antibody, intravenous immunoglobulin, ribavirin, or any other antiviral during the study. Study subjects had a typical duration of symptoms before hospitalization (or the first study encounter) of 4 days. Once hospitalized, the infants were generally enrolled into the study the following day, when their initial laboratory diagnosis of RSV infection was made. In general, the first symptom noted by the caregiver was a runny nose (data not shown); however, in many cases, the onset of the runny nose was on the same day as the onset of cough or even of the onset of difficulty breathing.
Quantitative virologic assessment.
The RSV loads showed a high degree of scatter, ranging from 1 to >6.78 log pfu/mL. The mean RSV load was 4.998 log pfu/mL (95% confidence interval [CI], 4.7645.232 log pfu/mL) (figure 1) and was normally distributed (Kolmogorov-Smirnov test). Because of the dilutions chosen for the plaque assay, the maximum detectable RSV load was 6.78 log pfu/mL. This value was reached in 28 (19.9%) of 141 subjects. The quantitative RSV standards, which were run in parallel with specimens from the subjects, showed excellent precision, with a mean ± SD of 5.000 ± 0.6328 log pfu/mL (95% CI, 4.8945.105 log pfu/mL).
Univariate analyses.
Subject factors and viral factors were analyzed to assess their ability to predict disease severity. Three individual markers of disease severity were analyzed: a requirement for intensive care, respiratory failure, and the duration of hospitalization. Univariate analyses revealed and confirmed several factors associated with increased disease severity (table 3). Lower weight on admission was the strongest predictor of all 3 markers of increased disease severity (P < .0001, for all 3 markers). Other factors, which are interrelated with weight on admission, were also significantly associated with increased disease severity. These were weight at birth, estimated gestational age, and age on admission. The traditional risk factors for RSV disease severitytobacco exposure, crowding (number of persons in the home), and absence of breast-feedingdid not correlate with any of the 3 markers of disease severity in the univariate analyses. The duration of symptoms before the collection of the nasal wash specimen was not related to disease severity, indicating that subjects who reportedly developed illness more rapidly did not statistically have more severe disease. White subjects had longer durations of hospitalization than did other subjects (4.78 vs. 3.80 days; P = .0099) but were no more likely to have respiratory failure or require intensive care. Day care attendees had less severe RSV disease than did nonattendees. This association likely reflects that caregivers purposefully avoided placing infants in day care who were at relatively higher risk for severe disease. The RSV load was higher in subjects requiring intensive care than in other subjects (5.194 vs. 4.949 log pfu/mL) and was also higher in subjects having respiratory failure than in other subjects (5.185 vs. 4.963 log pfu/mL), but these differences were not statistically significant in the univariate analyses. Similarly, subjects with higher RSV loads tended to have longer durations of hospitalization, but this trend was not statistically significant in the univariate analyses.
Analyses of independent effects.
Independent predictors of disease severity were analyzed using multiple regression and logistic regression approaches (table 4). An increased RSV load was a significant independent predictor of all 3 markers of disease severity. Furthermore, RSV load was the strongest independent predictor of disease severity, having the lowest conglomerate P value for the 3 separate severity markers. The statistical model allows for the magnitude of the effect to be quantified. The model predicts that, for every increase of 1 log pfu/mL in the RSV load, there will be an increase of 0.8 day in the duration of hospitalization. Both younger age on admission and lower weight on admission were independent predictors of more severe disease. However, of these, weight on admission was the stronger predictor. Other factors found to be associated with increased disease severity in the univariate model were not confirmed as independent predictors of disease severity. These included estimated gestational age and weight at birth. The absence of congenital anomalies (noncardiac) predicted shorter durations of hospitalization (P < .0001), but the occurrence of congenital anomalies was too infrequent to include it in the models that assessed the requirement for intensive care and respiratory failure. Reflecting its probable association with other risk factors, such as age and weight, nonattendance at day care was not an independent predictor of an increased duration of hospitalization (P = .132) or the requirement for intensive care (P = .131). Other factors that failed to independently predict any markers of RSV disease severity were sex, race/ethnicity, prematurity, breast-feeding, and tobacco exposure.
Effects on the RSV load.
We also analyzed, by univariate techniques, which factors affected the RSV load. Male subjects had 0.496 log pfu/mL (95% CI, 0.03800.9539 log pfu/mL; P = .0356) higher RSV loads than did female subjects. Race/ethnicity, tobacco exposure, breast-feeding, and presence of congenital anomalies did not affect the RSV load. The RSV load declined with an increasing duration of symptoms before the collection of the specimen (P = .0099) (figure 2). In aggregate, the slope of this decline was -0.13 log pfu/mL for each day of symptoms before enrollment into the study. Neither attendance at day care nor the number of children in the household correlated with the RSV load, indicating that these markers of crowding and possible exposure to a higher RSV inoculum could not be shown to affect the RSV load at this single time point. A higher weight at birth was associated with a higher RSV load (P = .0015). Similarly, subjects with higher weights on admission tended to have higher RSV loads (95% CI for the slope of the regression line, -85.66 to 397.7). Neither age on admission nor estimated gestational age correlated with the RSV load.
DISCUSSION
The main purpose of this study was to determine the independent effects that the RSV load has on disease severity. Here we have shown that the nasal aspirate RSV load, measured at an early time point during hospitalization, independently predicts clinically relevant markers of disease severity. We have previously shown that the nasal aspirate RSV load is associated with respiratory failure [18]. Here we have both confirmed this finding in a much larger data set and extended the association to include other markers of disease severity: the requirement for intensive care and the duration of hospitalization. We have also shown that other previously recognized predictors of severe RSV diseasesuch as lower age, congenital anomalies, and weight on admissionalso independently predict disease severity, thus helping to confirm the validity of the data set used in the present study.
We selected infants at an age at which they were most likely to be experiencing their first RSV infection. The RSV loads measured in parallel aliquots of respiratory secretions collected simultaneously by deep tracheal aspiration and by nasal wash are tightly correlated in infants at this age who are hospitalized [1921]. Thus, in our study population, nasal aspirate RSV loads are surrogate markers for pulmonary RSV loads at the time point examined. Our data suggest that RSV disease, although perhaps being largely a result of a pathogenic immune response, appears to be driven by the RSV load. Even though the RSV load is thought to be declining at the time of our single measurement [1921], an association with disease severity is still present.
Several previous attempts to associate disease severity and the RSV load have been unsuccessful [11, 2224]. Our ability to detect this association is most likely due to our sample size and study design [25]. We selected young subjects who most likely had not had a previous RSV infection. We also excluded those subjects who had conditions that might disrupt the putative relationship between RSV load and disease severity.
In the present study, the host factors that independently predicted disease severity were age, congenital anomalies, admission weight, and factors related to the timing of the disease. Prematurity and weight at birth were not independently associated with RSV disease severity but have been shown in several studies published elsewhere to be strong risk factors for severe RSV disease [1, 4, 17]. These factors were not independent predictors of disease severity in our study population, because we enrolled full-term infants. In our relatively gestationally mature infant population, it appears that the size of the airways (which is related to admission weight), the post-delivery maturity of the immune response, and/or lung structure/function (which is related to age) work in combination with viral load to affect RSV disease severity.
The present study is not without limitations. The plaque assay is subject to some inherent inconsistencies. To counteract this problem, all assays were performed by only 2 persons. Furthermore, the use of parallel quantitative standards ensured precise, reproducible results. Approximately 20% of the subjects had a measured RSV load above the threshold of quantification. This is likely to have affected the association between disease severity and RSV load. However, such an effect would reduce, not artificially augment, the ability to detect the association between RSV load and disease severity found in the present study. Although limiting the study to subjects who were, in general, without underlying disease allowed for a more efficient determination of the association between RSV disease severity and RSV load, this study design also may have limited the conclusions to ones that could apply only to this study population. It is likely that more severe RSV disease is produced by a higher RSV load even in infants with underlying disease. However, the magnitude of this association in such a population is likely different than that measured here.
Viral load, as measured in the present study, may be a function of 2 factors: the intrinsic growth characteristics of specific RSV isolates or subtypes and the host's ability to limit that growth. Because, at the time of our single measurement of the RSV load, the load is presumably declining [19, 21], it is likely that the capability of the infant's immune response to reduce viral replication is the major determinant of RSV load. Even so, viral factors not evaluated in the present study, such as genotype differences, may also predict disease severity. Such differences in RSV genotypes have been associated with disease severity [26], but this finding may not be universal [27].
Some studies that measured the concentrations of proteins or viruses in respiratory secretions have attempted to control for variations in the quality of the specimens by also measuring the concentration of an easily diffusible molecule, such as urea nitrogen, within those secretions [28, 29]. We elected not to use this technique, for several reasons. First, the differences in the concentrations of virus we expected to encounter were log differences, thus overriding the relatively small differences possibly caused by variations in the quality of the specimens. Second, we designed our study to control for the quality of the specimens at the time of the collection by quantifying and standardizing the collection procedure itself. Third, any differences in quality between the specimens from different subjects would cause random effects rather than systematic bias and, therefore, would not affect the observed statistically significant differences between population RSV load means. Last, we have previously conducted studies in similar groups of infected infants and measured RSV loads using the same collection techniques as in the present study and have demonstrated reproducible quantitative results [19, 21]. We have demonstrated that smooth RSV load curves are generated within individuals over time [1921], thus further validating these techniques in the absence of a control for the quality of the specimens after their collection.
In the univariate analysis, a higher weight at birth was significantly directly associated with a higher RSV load. In addition, a higher RSV load was directly associated with a higher weight on admission and a greater estimated gestational age, although these latter 2 associations were not statistically significant. These associations appear to be counterintuitive, but it is possible that nasal wash specimens obtained from larger subjects are collected from a larger surface area of respiratory epithelium and, therefore, may contain more virus than do nasal wash specimens obtained from smaller subjects.
Male sex has been associated with increased RSV disease severity [30]. The male : female ratio of infants hospitalized for RSV infection is almost universally >1. A slight predominance of males was also found in our study population. We found that RSV load was significantly higher (0.496 log pfu/mL) in male subjects than in female subjects. It has been proposed that sex-specific differences in the anatomy of the airways may be responsible for sex-based differences in the severity of viral respiratory tract infections such as RSV disease [31]. Differences in RSV load between the sexes may suggest another mechanism for the sex-based differences observed in the severity of this disease.
The timing and the magnitude of the effect that RSV load has on disease severity have potential therapeutic implications. For every decrease of 1 log in the RSV load, there was a reduction of 0.8 day in the duration of hospitalization. These data suggest that, if RSV load could be reduced by 1 log shortly after hospitalization, a reduction in the duration of hospitalization of almost 1 day may be achieved. Although this is possible, there are several reasons why such assumptions may be premature. First, it is possible that the RSV loads measured in the present study simply reflect higher RSV loads that occur earlier in the disease process. This earlier viral replication may be mainly responsible for driving disease which might not be reduced by subsequent later reduction in viral load. Immunogenetic differences are now being associated with RSV disease severity [3237]. Thus, there may be individuals with specific immune response genotypes/phenotypes that produce a more pathogenic immune response to RSV. These individuals may also have a limited ability to control the replication of RSV. Thus, viral load might be related to disease severity, but a reduction in viral load might not necessarily result in disease severity reduction. Antivirals do reduce the disease severity in other respiratory tract viral infections, but, to date, a compound with a robust antiviral effect has not been adequately studied as a treatment for RSV disease [38].
In conclusion, higher RSV loads predict greater disease severity in previously healthy naturally infected infants. This association holds true for several different clinically relevant markers of disease severity. The viral load now joins several other important host factors in predicting RSV disease severity. Whether relative inability to rapidly clear RSV predicts disease severity and whether the association between the RSV load and disease severity is strictly causal are questions that await further study.
Acknowledgments
Dr. DeVincenzo wishes to thank the infants and their parents, who volunteered to participate in this study; Malak Kotb and Elaine I. Tuomanen, for their mentorship in the conduct of this research; and Lisa Harrison and Jody Aitken, for their dedicated laboratory and clinical efforts.
References
1. Boyce TG, Mellen BG, Mitchel EF Jr, Wright PF, Griffin MR. Rates of hospitalization for respiratory syncytial virus infection among children in Medicaid. J Pediatr 2000; 137:86570. First citation in article
2. Leader S, Kohlhase K. Respiratory syncytial virus-coded pediatric hospitalizations, 1997 to 1999. Pediatr Infect Dis J 2002; 21:62932. First citation in article
3. Wang EE, Law BJ, Boucher FD, et al. Pediatric Investigators Collaborative Network on Infections in Canada (PICNIC) study of admission and management variation in patients hospitalized with respiratory syncytial viral lower respiratory tract infection. J Pediatr 1996; 129:3905. First citation in article
4. Wang EE, Law BJ, Stephens D. Pediatric Investigators Collaborative Network on Infections in Canada (PICNIC) prospective study of risk factors and outcomes in patients hospitalized with respiratory syncytial viral lower respiratory tract infection. J Pediatr 1995; 126:2129. First citation in article
5. Domachowske JB, Rosenberg HF. Respiratory syncytial virus infection: immune response, immunopathogenesis, and treatment. Clin Microbiol Rev 1999; 12:298309. First citation in article
6. Peebles RS Jr, Hashimoto K, Graham BS. The complex relationship between respiratory syncytial virus and allergy in lung disease. Viral Immunol 2003; 16:2534. First citation in article
7. Graham BS, Johnson TR, Peebles RS. Immune-mediated disease pathogenesis in respiratory syncytial virus infection. Immunopharmacology 2000; 48:23747. First citation in article
8. Kapikian A, Mitchell R, Chanock R, Shvedoff R, Stewart C. An epidemiologic study of altered clinical reactivity to respiratory syncytial (RS) virus infection in children previously vaccinated with an inactivated RS virus vaccine. Am J Epidemiol 1969; 89:40521. First citation in article
9. Graham BS, Bunton LA, Wright PF, Karzon DT. Role of T lymphocyte subsets in the pathogenesis of primary infection and rechallenge with respiratory syncytial virus in mice. J Clin Invest 1991; 88:102633. First citation in article
10. Alwan WH, Kozlowska WJ, Openshaw PJ. Distinct types of lung disease caused by functional subsets of antiviral T cells. J Exp Med 1994; 179:819. First citation in article
11. Legg JP, Hussain IR, Warner JA, Johnston SL, Warner JO. Type 1 and type 2 cytokine imbalance in acute respiratory syncytial virus bronchiolitis. Am J Respir Crit Care Med 2003; 168:6339. First citation in article
12. Polack FP, Teng MN, Collins PL, et al. A role for immune complexes in enhanced respiratory syncytial virus disease. J Exp Med 2002; 196:85965. First citation in article
13. Aung S, Rutigliano JA, Graham BS. Alternative mechanisms of respiratory syncytial virus clearance in perforin knockout mice lead to enhanced disease. J Virol 2001; 75:991824. First citation in article
14. Ohm-Smith MJ, Nassos PS, Haller BL. Evaluation of the Binax NOW, BD Directigen, and BD Directigen EZ assays for detection of respiratory syncytial virus. J Clin Microbiol 2004; 42:29969. First citation in article
15. Halstead DC, Todd S, Fritch G. Evaluation of five methods for respiratory syncytial virus detection. J Clin Microbiol 1990; 28:10215. First citation in article
16. DeVincenzo JP. Natural infection of infants with respiratory syncytial virus subgroups A and B: a study of frequency, disease severity and viral load. Pediatr Res 2004; 56:9147. First citation in article
17. Willson DF, Landrigan CP, Horn SD, Smout RJ. Complications in infants hospitalized for bronchiolitis or respiratory syncytial virus pneumonia. J Pediatr 2003; 143(Suppl 5):S1429. First citation in article
18. Buckingham S, Bush A, DeVincenzo J. Nasal quantity of respiratory syncytial virus correlates with disease severity in hospitalized infants. Pediatr Infect Dis J 2000; 19:1137. First citation in article
19. Malley R, DeVincenzo J, Ramilo O, et al. Reduction of respiratory syncytial virus (RSV) in tracheal aspirates in intubated infants using humanized monoclonal antibody to RSV F protein. J Infect Dis 1998; 178:155561. First citation in article
20. Malley R, Vernacchio L, DeVincenzo J, et al. Enzyme-linked immunosorbent assay to assess respiratory syncytial virus concentration and correlate results with inflammatory mediators in tracheal secretions. Pediatr Infect Dis J 2000; 19:17. First citation in article
21. Buckingham SC, Jafri HS, Bush AJ, et al. A randomized, double-blind, placebo-controlled trial of dexamethasone in severe respiratory syncytial virus (RSV) infection: effects on RSV quantity and clinical outcome. J Infect Dis 2002; 185:12228. First citation in article
22. Hall CB, Douglas RG, Geiman JM. Quantitative shedding patterns of respiratory syncytial virus in infants. J Infect Dis 1975; 132:1516. First citation in article
23. Hall CB, Douglas RG, Geiman JM. Respiratory syncytial virus infections in infants: quantitation and duration of shedding. J Pediatr 1976; 89:115. First citation in article
24. Wright P, Gruber W, Peters M, et al. Illness severity, viral shedding, and antibody responses in infants hospitalized with bronchiolitis caused by respiratory syncytial virus. J Infect Dis 2002; 185:10118. First citation in article
25. DeVincenzo JP, Buckingham SC. Relationship between respiratory syncytial virus load and illness severity in children. J Infect Dis 2002; 186:13767. First citation in article
26. Martinello RA, Chen MD, Weibel C, Kahn JS. Correlation between respiratory syncytial virus genotype and severity of illness. J Infect Dis 2002; 186:83942. First citation in article
27. Brandenburg AH, van Beek R, Moll HA, Osterhaus AD, Claas EC. G protein variation in respiratory syncytial virus group A does not correlate with clinical severity. J Clin Microbiol 2000; 38:384952. First citation in article
28. Rennard SI, Basset G, Lecossier D, et al. Estimation of volume of epithelial lining fluid recovered by lavage using urea as marker of dilution. J Appl Physiol 1986; 60:5328. First citation in article
29. DeVincenzo JP, Leombruno D, Soiffer RJ, Siber GR. Immunotherapy of respiratory syncytial virus pneumonia following bone marrow transplantation. Bone Marrow Transplant 1996; 17:10516. First citation in article
30. Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease. J Pediatr 2003; 143(Suppl 5):S11826. First citation in article
31. Martinez FD, Morgan WJ, Wright AL, Holberg CJ, Taussig LM. Diminished lung function as a predisposing factor for wheezing respiratory illness in infants. N Engl J Med 1988; 319:11127. First citation in article
32. Hacking D, Knight JC, Rockett K, et al. Increased in vivo transcription of an IL-8 haplotype associated with respiratory syncytial virus disease-susceptibility. Genes Immun 2004; 5:27482. First citation in article
33. Hoebee B, Bont L, Rietveld E, et al. Influence of promoter variants of interleukin-10, interleukin-9, and tumor necrosis factor genes on respiratory syncytial virus bronchiolitis. J Infect Dis 2004; 189:23947. First citation in article
34. Gentile DA, Doyle WJ, Zeevi A, et al. Cytokine gene polymorphisms moderate illness severity in infants with respiratory syncytial virus infection. Hum Immunol 2003; 64:33844. First citation in article
35. Hoebee B, Rietveld E, Bont L, et al. Association of severe respiratory syncytial virus bronchiolitis with interleukin-4 and interleukin-4 receptor polymorphisms. J Infect Dis 2003; 187:211. First citation in article
36. Choi EH, Lee HJ, Yoo T, Chanock SJ. A common haplotype of interleukin-4 gene IL4 is associated with severe respiratory syncytial virus disease in Korean children. J Infect Dis 2002; 186:120711. First citation in article
37. Hull J, Thomson A, Kwiatkowski D. Association of respiratory syncytial virus bronchiolitis with the interleukin 8 gene region in UK families. Thorax 2000; 55:10237. First citation in article
38. DeVincenzo JP. Therapy of respiratory syncytial virus infection. Pediatr Infect Dis J 2000; 19:78690. First citation in article