Objective. To determine the effect of respiratory viral infections on pulmonary function in infants with cystic fibrosis (CF) after the respiratory virus season (October through March).
Methods. Recruitment was for one respiratory virus season during a 3-year span, 1988 to 1991, with reenrollment allowed; 22 infants <2 years of age with CF (30 patient-seasons) and 27 age-matched controls (28 patient-seasons) participated. Primary outcome variables were preseason and postseason pulmonary function tests and serology for viral antibodies. Twice-weekly telephone calls screened for respiratory symptoms. The presence of respiratory symptoms triggered a home visit and an evaluation for upper or lower (LRTI) respiratory tract infection. A nasopharyngeal sample for viral culture was performed with each visit.
Results. Controls and CF infants each had a mean of 5.3 acute respiratory illnesses; CF infants were four times more likely to develop an LRTI compared with controls (odds ratio, 4.6; 95% confidence interval, 1.3 and 16.5). Three of 7 (43%) CF infants with respiratory syncytial virus infection (documented by culture) required hospitalization. Controls had no association between respiratory illness and postseason pulmonary function. For CF infants, reduced postseason maximal flow at functional residual capacity (V′maxFRC) was associated with two interactions, ie, respiratory syncytial virus infection and LRTI, and male sex and LRTI; increased gas trapping (FRC) was associated with an interaction between respiratory syncytial virus and LRTI and day care . Postseason pulmonary function tests were obtained a mean of 3.2 months after final LRTI.
Conclusions. Infants with CF incurring respiratory virus infection are at significant risk for LRTI, for hospitalization, and for deterioration in lung function that persists months after the acute illness.
- cystic fibrosis
- respiratory tract infections
- respiratory syncytial viruses
- respiratory function tests
Chronic obstructive lung disease continues to be the primary cause of morbidity in patients with cystic fibrosis (CF), despite significant improvement in survival during the past 20 years. Although children with CF have no detectable lung pathology at birth, half have acute or persistent respiratory symptoms at diagnosis.1 There is striking variability in disease progression and no clear association between specific cystic fibrosis transmembrane regulator protein gene mutations and severity of pulmonary disease.2 Determining the nongenetic factors that affect the progression of lung disease should prove important in the long-term treatment of children with CF.
Viral respiratory tract infections have been implicated in the development of progressive lung disease in CF. Respiratory viral infections cause significant lower respiratory tract disease in normal children younger than 5 years of age, whereas morbidity is substantially less in older children.3 One would expect a similar age-related pattern in the CF population. In older children and adults with CF, previous studies have demonstrated that respiratory viral infections are associated with pulmonary exacerbations, hospitalizations, or both.4–11
In the one reported study of respiratory viral infections in infants with CF,12 7 of 48 were hospitalized with proven respiratory syncytial virus (RSV) infection. Hospitalizations were prolonged, and on follow-up at a mean age of 26 months the 7 children had persistent chronic respiratory signs and lower chest radiographic scores compared with the other 41 children. Because infection was determined by culture only, the actual incidence of RSV infection among these patients is unknown. Pulmonary function was not assessed. The effects of respiratory viral infection on lung growth and development in infants with CF are therefore unknown.
To our knowledge the data presented here are from the first prospective study of respiratory viral infections and pulmonary function in infants with CF. The purpose of this study was to determine if, in infants (age, <2 years) with CF, respiratory viral infections were associated with reductions in pulmonary function after a 6-month respiratory virus season.
A prospective longitudinal study was conducted before, during, and after the 6-month respiratory virus seasons of 3 consecutive years, 1988 to 1991, with both CF patients and age-matched controls in each year's cohort. The infants were evaluated before and after the 6-month respiratory virus season (the season) and followed intensively during the season to distinguish upper (URTI) and lower (LRTI) respiratory tract infections and to identify the specific viruses associated with these illnesses.
Each year, enrollment was in the late summer and early fall, before the beginning of the season. Subjects were enrolled for one season at a time. A total of 26 infants with CF, age 1 to 20 months, were enrolled; 4 failed to complete the study and 8 others were reenrolled for a second season, for a total of 30 CF patient-seasons. A total of 27 age-matched controls were enrolled; all completed the study and 1 reenrolled for a second season, for a total of 28 patient-seasons. To facilitate assessment of acute respiratory viral illnesses, all recruited subjects lived within 50 miles of the sponsoring institution. The infants with CF were recruited through the Cystic Fibrosis Center at Baylor College of Medicine. Diagnosis of CF followed the standards of the Cystic Fibrosis Foundation and was based on the presence of two or more of the following criteria: (1) sweat chloride ≥60 mEq/L based on a repeated quantitative pilocarpine iontophoresis sweat test; (2) genetic testing demonstrating homozygosity for the ΔF508 allele; and (3) clinical features consistent with CF. The age-matched controls were recruited from a private pediatrician's office and by an advertisement in local newspapers; each had a negative history of croup, bronchiolitis, chronic cough, asthma, and prematurity. The study was approved by the Institutional Review Board of Baylor College of Medicine and Texas Children's Hospital, and signed informed consent was obtained from at least 1 parent.
At enrollment, subjects in both groups (controls and infants with CF) had a physical examination. For both groups, demographic data, family medical history, the infant's medical history, and environmental risk factors were documented through questionnaires; the questionnaire regarding environmental risk factors was repeated at the end of the season. Pulmonary function tests (PFTs), throat cultures, and serology for viral antibodies were performed at entry into the study and 6 months later at the end of the season (preseason and postseason). The PFTs included oxygen saturation by pulse oximetry (Spo2), maximal flow at functional residual capacity (V′maxFRC), functional residual capacity (FRC), and respiratory rate. Throat cultures were also obtained preseason and postseason from both groups to determine the presence of pathogenic bacteria, specifically Staphylococcus aureus andPseudomonas aeruginosa. At the same time, antibody levels to RSV group A, adenovirus, influenza A and B viruses, and parainfluenza virus type 3 (PIV-3) were assayed. For the infants with CF, Brasfield radiographic and modified Shwachman clinical scores were obtained preseason and postseason. The infants with CF received a trivalent influenza vaccination at the preseason evaluation, and an additional blood sample for antibody titers was obtained 4 weeks later. Controls were not immunized against influenza.
The respiratory virus season was defined by using sentinel cases of RSV as monitored by the Influenza Research Center at Baylor College of Medicine in Houston, Texas. The Research Center monitored several clinics throughout the metropolitan area and obtained cultures from a sample of adults and children diagnosed with respiratory viral infections. The first RSV-positive culture each fall marked the beginning of the season, which lasts an average of 6 months. In the Houston area, the RSV epidemic normally starts in October and peaks in December or January; this is earlier than in many other parts of the country.
The study coordinator was either a nurse (2 years of the study) or a physician (1 year). Before patient recruitment, the study coordinator was trained in the evaluation and assessment of URTIs and LRTIs in both normal and CF infants. Throughout the season, the study coordinator made telephone contact twice each week with the families of both the CF and the control infants, to monitor for any symptoms of respiratory illness. When an acute respiratory illness was identified, the study coordinator made a home visit within 24 hours. The coordinator evaluated the infant and, using a modified respiratory illness score,11,,13 determined if the illness was an acute URTI or LRTI, and obtained a nasopharyngeal wash/throat swab specimen, which was cultured for respiratory viruses. The results of the clinical assessment were given to the infant's primary care provider, who determined treatment.
CF Clinical Scores
For the infants with CF, chest radiographs were evaluated preseason and postseason by the Brasfield scoring system (25 = best, 4 = worst).14 The severity of each infant's CF symptoms was quantified preseason and postseason by using a respiratory/nutritional score developed for children with CF (the modified Shwachman clinical score: 75 = best, 4 = worst).15 Measurements of height and weight, adjusted for age and sex, and symptoms of gastrointestinal involvement (eg, descriptions of stools) are included in the modified Shwachman score.
The demographic questionnaires included the information for determining socioeconomic status (SES), using the Hollingshead score,16 in addition to standard data regarding race, sex, and number and ages of siblings.
For both CF patients and controls, the PFTs included Spo2, V′maxFRC, FRC, and respiratory rate. Infants were free of respiratory viral illness for a minimum of 2 weeks before lung-function testing. Those who were being treated with bronchodilators had their medications withheld for at least 4 hours before testing. A pulse oximeter (Nellcor N-200; Nellcor, Inc, Hayward, CA) measured Spo2 from the great toe or finger; Spo2 was monitored continuously and recorded once a stable value had been demonstrated for at least 1 minute. The infant received 65 to 75 mg/kg of chloral hydrate orally; pulmonary function was then assessed in the supine position. Lung volumes at FRC were measured by the helium dilution closed-circuit technique, as previously described.17,,18Changes in the circuit volume and helium concentration versus time were recorded on an eight-channel recorder (Beckman; Schiller Park, IL) and each FRC measurement was completed when the helium concentration remained constant for 30 seconds. Tidal volume and FRC were calculated from the recordings of the volume and helium concentration. Partial expiratory flow volume curves generated by the rapid chest compression technique18 were used to measure V′maxFRC. Forced expiratory maneuvers were repeated at increasing compression pressures until the maximal value was obtained and then repeated in triplicate at the optimal compression pressure. In all cases, the three maximum flow measurements were within 10% of each other; the highest measurement of V′maxFRC was used for data analysis. Sleeping respiratory rate was measured over a period of 1 minute and served as the infant's baseline respiratory rate.
Respiratory Illness Score
No infant with CF had chronic findings of pulmonary disease (crackles or wheeze) at enrollment. Kenner's respiratory illness score, as adapted by Wang, was used at home visits to distinguish acute URTIs and acute LRTIs.11,,13 An acute URTI was diagnosed if the infant had sneezing, coryza (rhinorrhea, nasal congestion, or nasal crusting), or pharyngitis (pharyngeal erythema or exudate). The criteria for a LRTI were (1) wheeze or crackles on auscultation of the lungs; (2) shortness of breath; (3) a respiratory rate of >15 breaths/min above baseline respiratory rate; (4) increase in their normal frequency of cough; and (5) increase in sputum production or a change in the quality of the sputum (from clear to turbid yellow or green). As children in this age group do not expectorate, the sputum criterion was not used in this study. All infants assessed as having an acute LRTI had new findings of crackles or wheeze on auscultation, and many also showed shortness of breath, an increased respiratory rate, or an increase in cough.
Infection with specific respiratory viruses was documented by isolation of the virus in tissue culture, by measuring a fourfold or greater rise in postseason serum antibody titer compared with preseason titers, or by both culture and titer. Respiratory viruses were isolated from nasopharyngeal samples using HEp-2 (human laryngeal tumor cells), MDCK (Madin-Darby canine kidney cells), LLC-MK2 (Lewis lung carcinoma—monkey kidney [cell culture]), and Wister Institute human diploid lung cell line (WI-38 cell cultures19); these samples were also tested for RSV antigen by rapid antigen detection test according to the manufacturer's instructions (Directogen, Becton Dickinson, Cockeysville, MD). Samples to be cultured were generally transported to the laboratory within 6 hours and usually were not frozen. Broad-spectrum antimicrobials (pipercillin, gentamicin, vancomycin, and amphotericin B) were added to the viral cultures to prevent overgrowth by bacteria and fungi commonly found in CF patients. Microneutralization assays were used to determine serum antibody titers to influenza A and B, PIV-3, and RSV group A viruses.20–22 Antibodies directed to the antigenic determinants (group-specific and type-specific) of adenoviruses were measured by using an enzyme-linked immunosorbent assay. The purified hexon protein of the adenovirus type 5 was the antigen used in the enzyme-linked immunosorbent assay.23
Comparisons of the distributions of categorical variables between groups were examined by χ2 or Fisher's exact test and the means of continuous variables by two-tailed Student'st test. Nutritional indices were adjusted for age and sex with CDC ANTHRO Software for Calculating Pediatric Anthropometry, Version 1.01 for Windows (Atlanta, Ga).
The data are presented for 30 CF patients and 28 normal age-matched controls, except for the regression analyses. Using >1 year of data for the subjects who reenrolled for a second season would have violated the assumption of independence for multiple regression. The regression analyses evaluate 1 year's data for each subject. As RSV infection was a major interest, the data used were from the year in which an RSV infection was documented.
Occurrence of an LRTI was examined as a predictor of postseason V′maxFRC by using multiple linear regression techniques. The independent variables included baseline characteristics (age, sex, ethnicity, weight/length Z score, initial Brasfield radiographic score), environmental factors (SES score and day care), and respiratory viral infection and change in length (growth) during the 6 months. When two independent variables were highly correlated, the variable with the most clinical significance (eg, length rather than age) was chosen for the model. Through a process of backward elimination, independent variables and interaction terms that did not contribute to the model were omitted. Occurrence of an LRTI was also examined as a predictor of postseason FRC, using the same modeling procedure.
In a secondary analysis, the infants with CF who had one or more LRTIs during the season (the CF LRI group) were compared with those who had only URTIs (CF URI-only group). Comparison of their baseline measures followed the procedure used for the comparison of the CF and control groups. Differences in antibody responses to the trivalent influenza vaccine between the CF LRI and the CF URI-only groups were evaluated in separate linear regression models with postvaccine titers to influenza A virus, either H3N2 or H1N1, or to influenza B virus as the dependent variable. The corresponding baseline titer, age, and weight/lengthZ scores were entered as independent variables.
A significance level of P < .05 was used throughout. The analyses were performed with Statistical Analysis System software for Windows (SAS Institute, Inc, Cary, NC).
Primary Analysis: Controls and CF Patients
There were no significant differences between the infants with CF and the normal controls for family history of asthma, exposure to environmental tobacco smoke, age, sex, or length at entry into the study (Table 1). The number and ages (0–5 years, 6–10 years, and 11–18 years) of other children in the home were similar, but the control subjects had a higher SES (P < .01). Weight and weight/length Zscores were significantly lower for the infants with CF (P < .01). The infants with CF also had impaired pulmonary function; ie, their V′maxFRC/length was worse (P < .05), their respiratory rates were higher (P < .01), and their Spo2value was lower (P < .01). Controls and infants with CF had similar low antibody titers to the viral agents that were measured, except to PIV-3, which was higher in the controls at entry into the study (P < .05).
Controls and infants with CF had similar numbers of acute respiratory tract illnesses during the 6-month respiratory viral season, every child incurring at least one acute URTI. However, the rate of lower respiratory tract involvement was significantly different for controls and infants with CF (Table 2). The infants with CF were four times more likely to develop an acute LRTI (odds ratio, 4.6; 95% confidence interval, 1.3 and 16.5; P < .05). None of the controls were hospitalized for LRTI. Of the infants with CF, 7 were hospitalized for acute LRTIs, 1 on two separate occasions. During these eight hospitalizations, RSV was cultured in 3 patients and picornavirus in 1 patient.
Controls showed twice as many documented infections (by culture, titer, or both) with RSV, influenza viruses A and B, PIV-3, and adenovirus over the course of the season (Table 2). This difference was statistically significant (P < .01) for RSV and approached significance for influenza viruses A and B (P < .07). Virus was isolated by culture significantly more frequently in the control group, ie, 40 of 141 (28%) nasopharyngeal specimens compared with 26 of 150 (17%) specimens from the infants with CF (P < .05).
Infection and Lung Function
Two primary regression analyses were conducted, one using postseason V′maxFRC as the dependent variable and the other using postseason FRC. For the age-matched controls, the regression analyses demonstrated that postseason V′maxFRC was associated with higher flows at the beginning of the study and with sex. In a similar manner, postseason FRC was associated with baseline FRC and with sex. For the age-matched controls, there was no association between changes in lung function and LRTI or any specific respiratory virus infection.
Table 3 shows the results of these two regression analyses for the infants with CF. In the first regression analysis, with postseason V′maxFRC as the dependent variable, two significant interactions were found; ie, V′maxFRC was lower in infants who had been infected with RSV and had incurred one or more LRTIs (P = .02) and in males who had one or more LRTIs (P < .001). It was also lower in those with low household SES scores (P < .01). In the second regression analysis, using FRC as the dependent variable, one significant interaction was observed, ie, RSV infection and one or more LRTIs (P = .01). Adenovirus infection (P = .01) and enrollment in day care (P= .03) were also significantly associated with worsening lung function (increased FRC).
Secondary Analysis: CF LRI and CF URI-Only
The result of the first regression analyses prompted us to compare the baseline characteristics, before the occurrence of any respiratory infections, of the 13 infants with CF who later had an LRTI (CF LRI) and the 17 who had only URTIs (CF URI-only). As shown in Tables 4 and 5, the significant differences at the preseason evaluation (baseline) between these two groups were sex, weight/length Z scores, radiographic and clinical scores, and titers to RSV group A (P < .05). The infants in the CF LRI group also had more symptoms of malabsorption (loose, frequent stools and steatorrhea;P < .05) at baseline, compared with the infants in the CF URI-only group. In the postseason evaluations, Brasfield scores were 21 ± 2 (CF URI-only) and 21 ± 2 (CF LRI) (P= .3) and the modified Shwachman scores were 86 ± 14 (CF URI-only) and 78 ± 10 (CF LRI) (P = .09). The improvement in Shwachman scores was primarily nutritional. The proportion of newly diagnosed (enrolled within 6 weeks of diagnosis) was similar in the two groups (5 of 17 CF URI-only and 5 of 13 CF LRI).
To evaluate possible differences in immune responses between the CF URI-only and CF LRI groups, antibody response to the trivalent influenza vaccine was evaluated (Table 6). After immunization, a fourfold or greater rise in antibody titer occurred more frequently in the CF URI-only group compared with the CF LRI group. Influenza B titers showed a significant difference (P = .012), and there was a strong trend with respect to influenza A virus H1N1 (P = .056). Postvaccine titers to influenza A H3N2, H1N1, and influenza B were used as the dependent variables in separate linear regression models (not shown). Antibody response appeared to be closely related to baseline antibody titers, occurrence of LRTI, and nutrition; ie, infants with CF who had higher baseline antibody titers had better antibody responses (P < .05), whereas infants with CF who had LRTIs during the respiratory season and those with poorer nutritional status at baseline (lower weight/length Z scores) had antibody responses to the vaccine that were significantly lower (P < .05).
Infection and Lung Function
Figure 1 illustrates the mean changes ± SEM in V′maxFRC, preseason to postseason, for the normal controls, for the CF LRI group, and for the CF URI-only group. For the CF LRI group, the postseason measures were obtained a mean of 3.2 months after the final LRTI (as determined by the respiratory illness score). As indicated in Table 1, controls had higher flows at the beginning of the study compared with infants with CF (P < .05). There was no significant difference in the flows of infants in the two CF subpopulations (CF URI-only and CF LRI) at the beginning of the study (Table 5). Comparing these data with those obtained at the end of the study, infants in the CF URI-only group showed a significant gain in V′maxFRC/length over the course of the 6 months (P < .05). Infants in the CF LRI group had a decrease in V′maxFRC/length; the difference in postseason flows between infants with CF who did and did not incur an LRTI was significant (P < .05).
The infants with CF and the normal controls had the same number of respiratory illnesses over a single respiratory virus season; the outcomes of those infections, however, were sharply different. In infants with CF, a viral respiratory infection was significantly more likely to become an acute infection of the lower respiratory tract, impair pulmonary function, and result in hospitalization; ie, almost 1 in 4 were admitted to the hospital for LRTI. Our data also showed that viral respiratory infections in infants with CF were associated with decreased lung function for several months (mean, 3.2 months) after LRTI.
Impact of Viral Infection on Pulmonary Function
Our normal control infants increased their pulmonary function over the 6-month period, as expected with growth. All the controls had at least one acute URTI, and 4 had one or more acute LRTIs as well. In the regression analysis of the control group, LRTI was not associated with any decrease in postseason pulmonary function. In contrast, the regression analyses (Table 3) show that in infants with CF, RSV and other lower respiratory tract infections were strongly associated with a deterioration in V′maxFRC, despite as long as 5 months elapsing between the last documented LRTI and the postseason PFT. Adenovirus and RSV infections were associated with increased airway obstruction past the acute phase of the illness, as shown by increasing FRC. The association of LRTIs with worsening flows (V′maxFRC), and of adenovirus and RSV with increased gas trapping (FRC), appears to confirm the clinical experience that LRTIs have a negative impact on the clinical course of lung disease in infants with CF.
Timing of the PFTs may be a potential confounding factor; the 2-week minimum time between respiratory symptoms and evaluation of lung function was a practical compromise, as children in this age group average six or more respiratory illnesses per year. The actual elapsed time between the last viral respiratory illness (URTI or LRTI) and the postseason PFT was a mean of 5 weeks. A mean of 3.2 months elapsed between the last LRTI and the postseason PFT, which supports the inference that the results indicate a longer-term impact of LRTI on pulmonary function rather than the effects of an acute illness.
Hospitalization for RSV LRTI
Seven of the infants with CF had RSV infections confirmed by titer, culture, or both (Table 2), and 3 of the 7 (43%) were hospitalized for acute respiratory distress. None of the age-matched controls were hospitalized for LRTI; among normal children this age, the reported hospitalization rate for RSV LRTI is typically 1% annually.24,,25 It is possible that physicians are more likely to hospitalize infants with CF than previously healthy infants presenting with similar respiratory symptoms. A total of eight hospitalizations for LRTI were recorded among the infants with CF (Table 2; 1 child was admitted twice); the admitting physicians were the infants' primary CF-care providers and were not study investigators. Review of the medical records showed that all eight admissions were prompted by respiratory distress. Hospital stays ranged from 3 to 21 days (mean, 9 days); the length of stay suggests severe lower respiratory tract disease. Although the sample size is small, the risk for hospitalization with RSV infection would appear to be substantially greater for infants with CF.
Identification of Viruses and Documentation of Infection
We had a 17% positive isolation rate by viral culture for the infants with CF (Table 2), comparable with published isolation rates with cultures from normal children.26 Approximately one-third of the published studies of respiratory viral infection in patients with CF have not attempted culture of the virus during acute illnesses, and most of the rest have been unable to consistently isolate viruses by culture. Our greater success may be because of our use of antibiotics in the tissue culture media, to the rapid processing of most specimens, or to the young age of the participating CF patients. A higher concentration of virus can be recovered from the nasal secretions of children, compared with adults, in the normal population; whether this holds true for the CF population is not known.
Despite this success, we may still have failed to identify some of the respiratory viruses infecting the infants with CF; the 17% identification rate in these patients was substantially lower than the 27% isolation rate for our normal controls. However, serologic analysis also showed a lower incidence of respiratory viral infection in the infants with CF. This suggests that the difference in viral culture represents a difference in viral infection rates. The comparable preseason neutralizing antibody titers of the control and CF groups (Table 1) would indicate a similar risk of infection on exposure, but the controls had higher rates of documented infection for every virus studied (Table 2). The difference in documented influenza infection (Table 2) was probably the result of the influenza immunization administered to the CF infants who were 6 months old or older. With respect to the other viruses studied, the difference in documented infection rates may reflect the significant difference between the CF patients and the controls in day care enrollment. Day care is associated with higher rates of respiratory viral infection in the general population because of the associated increase in exposures. The rate of participation in day care (Table 1) was 31% for controls and 10% for the CF patients. In contrast, SES was also significantly higher for the controls, and lower SES is associated with more crowded living conditions and greater ease of transmission of respiratory viruses. Limited sera prevented analysis for additional microorganisms, such as Mycoplasma, that might be of interest.
Positive nasopharyngeal cultures for P aeruginosa have a high predictive value (83% in Ramsey's study27) for lower airway colonization in children with CF. The preseason throat cultures indicated that >50% (16 of 30) of the infants with CF were colonized with P aeruginosa (Table 4). Studies of older CF children have shown that those colonized with P aeruginosatend to have greater morbidity with respiratory virus infections.8 This effect was not observed in the regression analysis in our study.
Differences in Immune Response
As part of our secondary analysis of possible differences between the CF URI-only and CF LRI groups, we evaluated the differences in the responses of those 6 months old and older to trivalent influenza vaccine. We found that those in the CF LRI group had significantly lower serum antibody responses to the three influenza virus antigens (Table 6). Regression analysis showed three factors predicted antibody response. Two have been reported in the literature, ie, age and baseline antibody titers. Those who were younger mounted a poorer antibody response, an observation consistent with published studies of both normal and CF infants and children given inactivated vaccines.21,,28 Higher baseline antibody titers correlated with higher postvaccine titers. This association suggests priming of the immune system, either by previous infection or by influenza immunization.
The third factor affecting antibody response was weight/lengthZ score; infants with lower weight/length Zscores, indicating they were less well-nourished, mounted a poorer antibody response to the influenza vaccine. At the same time, the infants in the CF LRI group had, at entry into the study, a history of more severe malabsorption compared with the CF URI-only group. In other populations, increased morbidity and mortality with viral respiratory infections have been highly correlated with the degree of malnutrition.29 It is possible that malabsorption-linked malnutrition may be a risk factor for the severity of respiratory viral infection in infants with CF. Our study was not designed to determine the effect of protein-calorie malnutrition on respiratory viral infections in infants with CF, but further detailed investigation may be appropriate.
Compared with normal controls, infants with CF appear to have more lower respiratory tract involvement with respiratory virus infection, an increased risk of hospitalization, and deterioration in lung function after the acute illness. Reducing the number of LRTIs may prevent deterioration in pulmonary function in these patients. Malnutrition may have an impact on the severity of respiratory viral illness and thus indirectly affect pulmonary function in this population. Children with CF should be considered prime candidates for RSV prevention trials using RSV intravenous immunoglobulin, RSV monoclonal antibodies, and RSV subunit vaccines.
This study was supported by a grant from the Cystic Fibrosis Foundation.
We thank Karyn Popham (Houston, Texas) for her editorial assistance.
- Received April 21, 1998.
- Accepted August 20, 1998.
Reprint requests to (P.W.H.) Baylor College of Medicine, Pediatric Pulmonology, 6621 Fannin, MC 3–2571, Houston, TX 77030.
- CF =
- cystic fibrosis •
- RSV =
- respiratory syncytial virus •
- URTI =
- upper respiratory tract infection •
- LRTI =
- lower respiratory tract infection •
- PFT =
- pulmonary function test •
- Spo2 =
- oxygen saturation by pulse oximetry •
- V′maxFRC =
- maximal flow at functional residual capacity •
- FRC =
- functional residual capacity •
- PIV-3 =
- parainfluenza virus type 3 •
- SES =
- socioeconomic status •
- CF LRI =
- group of infants with CF who had one or more LRTIs during the season •
- CF URI =
- group of infants with CF who had only URTIs
- Accurso FJ
- Tsui L,
- Durie PR
- ↵Hall CB. Respiratory syncytial virus. In: Feigin RD, Cherry JD, eds. Textbook of Pediatric Infectious Diseases, II. 2nd ed. Philadelphia PA: WB Saunders Company; 1998:2084–2111
- Efthimiou J,
- Hodson ME,
- Taylor P,
- Taylor AG,
- Batten JC
- Ong EL,
- Ellis ME,
- Webb AK,
- et al.
- Przyklenk B,
- Bauernfeind A,
- Bertele RM,
- Deinhardt F,
- Harms K
- Stroobant J
- Brasfield D,
- Hicks G,
- Joong S,
- Tiller RE
- ↵Hollingshead AB. Four Factor Index of Social Status (working paper). New Haven, CT: Yale University, Department of Sociology; 1975
- Scott JV,
- Dreesman GR,
- Kasel JA
- Groothuis JR,
- Levin MJ,
- Rabalais GP,
- Meiklejohn G,
- Lauer BA
- Copyright © 1999 American Academy of Pediatrics