- Email: [email protected]
The effect of ribavirin to treat previously healthy infants admitted with acute bronchiolitis on acute and chronic respiratory morbidity
RESPIRATORY MEDICINE (2001) 95, 275–280 doi:10.1053/rmed.2001.1033, available online at http://www.idealibrary.com on
The effect of ribavirin to treat previously healthy infants admitted with acute bronchiolitis on acute and chronic respiratory morbidity M. L. EVERARD*, A. SWARBRICK*, A. S. RIGBY{ AND A. D. MILNER* *Paediatric Respiratory Unit, Queens Medical Centre, Nottingham, U.K. and {Division of Child Health, Sheffield Children’s Hospital, Sheffield, U.K. The role of ribavirin in the treatment of acute bronchiolitis is controversial. It has been suggested that the use of ribavirin may be of benefit during the acute illness and may reduce subsequent recurrent respiratory morbidity. This randomized, double-blind, placebo-controlled study was designed to determine whether ribavirin administered during the acute illness would have an influence on respiratory morbidity during both the acute illness and during the following year. Bronchial reactivity 6 months after the acute illness was also assessed. Forty previously well infants with moderately severe acute bronchiolitis were recruited during three winter epidemics. Subjects received study medication for 18 h a day. Management was otherwise unaltered. Subjects were evaluated daily by the investigator and subsequently assessed at 6 weeks, 6 months and 1 year following the acute illness. Assessment of bronchial hyper-responsiveness was assessed at 6 months of age using total body plethysmography and an established ultra-sonically nebulized distilled water challenge. A total of 40 patients (21 ribavirin, 19 placebo) were entered into the study. The two groups did not differ with respect to age, gender or clinical severity on entry to the trial. No significant differences were identified in the rate of clinical improvement over the first 24 h, the time to discharge, bronchial responsiveness at 6 months of age, frequency of significant respiratory symptoms over the first year of life and the frequency of prescribed bronchodilators and inhaled steroids during the year of follow-up. This study was unable to demonstrate any clinical benefit from the use of ribavirin in the acute illness or during subsequent follow-up for 1 year. Key words: respiratory syncytial virus; acute bronchiolitis; ribavirin; respiratory morbidity. RESPIR. MED. (2001) 95, 275–280
Introduction Acute bronchiolitis is the commonest form of lower respiratory tract infection in infancy, with 0?5–2% of all infants being hospitalized with this condition during the winter following their birth (1–3). In the majority of cases the causative organism is the respiratory syncytial virus (RSV). For the vast majority of infants admitted with this condition, careful monitoring and good supportive care remains the cornerstone of management (4,5). It has been suggested that the anti-viral drug, ribavirin, has a role in the treatment of these infants but despite the fact that ribavirin has been licensed for the treatment of infants with respiratory syncytial virus for more than a decade, its place in the treatment of RSV bronchiolitis remains controverReceived 15 September 2000 and accepted in revised form 2 January 2001. Correspondence should be addressed to: Dr Mark L. Everard, Paediatric Respiratory Unit, Sheffield Children’s Hospital, Western Bank, Sheffield S10 2TH, U.K. E-mail: [email protected]
0954-6111/01/040275+06 $35?00/0
# 2001 HARCOURT PUBLISHERS LTD
sial. This has been reflected in the position of the American Academy of Pediatrics recent reassessment of the indications for its use which resulted in the recommendation that practitioners should be ‘allowed to decide whether ribavirin therapy is appropriate by taking into account the particular clinical situation and their own preferences’ (6). Previous randomized double-blind studies have concentrated on the effect of this drug on the course of the acute respiratory illness. However, the morbidity associated with RSV bronchiolitis is not confined to the acute illness in that the majority of infants admitted to hospital with acute bronchiolitis will subsequently experience recurrent lower respiratory tract symptoms, particularly in the pre-school years (7–9). The cause of these recurrent respiratory symptoms is unclear (4,10). It may be that certain forms of pre-existing respiratory dysfunction predispose individual infants to both acute bronchiolitis and to the subsequent recurrent symptoms. Alternatively, changes in the lower airway structure or physiology caused by the intense inflammation present in the airways of infants with acute bronchiolitis (11) may contribute to the subsequent # 2001 HARCOURT PUBLISHERS LTD
276
M. L. EVERARD
ET AL.
respiratory morbidity. Other suggestions have included the suggestion that an RSV-specific IgE response contributes to subsequent morbidity (12), although recent studies suggest that this is unlikely (13,14). Previous studies have provided convincing evidence that RSV bronchiolitis does not promote expression of atopy (4,7), although a recent study suggests that this may be the case (15). If recurrent symptoms are attributable to the effects of virally-induced inflammation, then it is possible that the use of ribavirin early in the illness would limit viral replication, limit the intensity of the inflammatory process, and so reduce the frequency of post-bronchiolitic symptoms. This randomized, controlled, double-blind study was designed to determine whether administering ribavirin to previously healthy infants admitted with acute bronchiolitis would reduce the frequency of lower respiratory tract symptoms in the year following admission. In addition, we studied the effect of ribavirin therapy on the course of the acute illness and assessed bronchial responsiveness at 6 months following the acute illness using a nebulized distilled water challenge.
Methods Forty infants admitted to the paediatric wards of the Queens Medical Centre, Nottingham with moderately severe bronchiolitis were recruited during three RSV epidemics. All were previously fit and had no high risk factors for severe disease. All were entered into the study within 12 h of admission. To enter the study the definition of ‘bronchiolitis’ used in the U.K. and Australia was used (16,17), that is, patients with evidence of upper respiratory tract viral illness who develop lower respiratory tract involvement characterized by airways obstruction and widespread crepitations on auscultation. Wheeze and rhonchi may or may not be present. Patients were entered into the study before their RSV status was assessed by an immunofluorescence technique to ensure that they were entered into the study as soon after admission as possible in line with the suggestion that the early use of ribavirin is associated with the greatest benefit. These patients were randomized to receive 6 g of ribavirin in 180 ml of water by the Small Particle Aerosol Generator (SPAG). The medication was administered following a standard regime over 18 h per day. Normal saline was used for the placebo since previous studies had been criticised for using distilled water as placebo. It has been suggested that nebulized distilled water may adversely affect these infants by inducing bronchoconstriction (18). However, the output of the SPAG is so low that this seems unlikely (19). The study was approved by the local Ethics Committee and informed consent was obtained from the parents of all subjects.
ACUTE ILLNESS Subjects were evaluated daily until discharge and their clinical status was assessed by one investigator (M.L.E.), who undertook assessments in a blinded fashion. In order
to ensure that the assessment was blinded the SPAG was switched off and any crystalline ribavirin removed from tubing, bedding and patient before the patient was assessed. Patients were assigned a daily severity score. No scoring that accurately reflects the severity of the acute illness has been validated. The arbitrary scoring system chosen used a scale of 1–10; 1 representing no symptoms and 10 would be applied to a ventilated infant. Time until discharge from hospital was recorded with an additional outcome of ‘fit for discharge’ being recorded. ‘Fit for discharge’ was defined as time when patients were deemed to be medically fit for discharge. Time to discharge can be influenced by social factors and this second parameter was included to try and ensure social factors did not obscure treatment effects. Duration of supplementary oxygen were all recorded.
FOLLOW-UP Following discharge, subjects were reviewed at 6 weeks, 6 months and 1 year after discharge. Patients were provided with diaries at each visit in which they were asked to record all respiratory symptoms, medical attention for any respiratory symptoms and any medication prescribed for respiratory symptoms. The dates of onset and resolution of symptoms were also recorded. Parents were asked to record any of the following respiratory symptoms—running nose, cough, wheeze and shortness of breath. Bronchial hyper-responsiveness was assessed using an ultra-sonic distilled water challenge as previously established in our laboratory (20). A number of approaches to assessing bronchial hyper-responsiveness in infants using total plethysmography have been described using pharmacological agents (21–23). However we chose to use a ‘nonspecific’ hypo-osmolar challenge as it was felt that this may be more clinically relevant than using a mediator such as histamine. This approach has been used by a number of centres studying bronchial responsiveness in adults and older children. The use of an ultra-sonic distilled water challenge was evaluated in 100 infants prior to the commencement of this study (20). A positive response to the ultra-sonic challenge was defined as a fall in specific airways conductance (sGaw) of greater than 20% from baseline based on previous work assessing reproducibility of measurements (20). Airways conductance was assessed using absolute total body plethesmography which had been used to assess airways conductance for more than four decades (24) and was used in our unit for more than a decade prior to this study. The technique has been used extensively over recent years, providing insights into a wide variety of cardiopulmonary disease and response to therapy (24). The principles and practical considerations have been recently been reviewed in considerable detail (24). The total body plethysmograph used in this study was a purpose built unit constructed by the Department of Medical Physics, Queen’s Medical Centre, Nottingham based on a standard design (20). The volume of the plethysmograph was 65 l. Once sedated, with 100 mg kg71 of chloral hydrate, infants were placed in the plethysmograph and allowed to settle. Baseline
RIBAVIRIN: ACUTE AND CHRONIC RESPIRATORY MORBIDITY measurements were made when the system had equilibrated and there were minimal changes (less than 0?58) in the slope of the box pressure/slope readings from the face of the oscilloscope over a 5-min period. Infants then inhaled ultrasonically nebulized distilled water for 2 min administered via an Intersurgical Variosonic nebulizer, with the aerosol being delivered via the tube connecting the infants face mask to the outside of the box ensuring that the box and infant were not disturbed. Measurements of sGaw were made at 2-min intervals for a total of 12 min after the nebulization was completed.
STATISTICS The ribavirin and placebo groups were analysed by comparison of the means: 95% confidence intervals were calculated for the mean differences. The results presented in Table 3 give both unadjusted and adjusted mean differences (adjusting for age at admission, sex and weight of baby). The adjusted analyses were carried out by linear leastsquares regression method on each of five outcome measures (i.e. change in score, days in oxygen, changes in oxygen saturation measured in air, days to discharge and days fit for discharge). For each regression analysis, the residuals followed an approximate normal distribution. A nominal significance level of 5% was assumed throughout.
Results Demographic details of the patients are included in Table 2. The two groups did not differ in respect to age, weight or
TABLE 1. Lower respiratory tract symptoms and therapy required during the first year following admission to hospital with acute brionchiolitis
Admitted with LRT symptoms Bronchodilators Inhaled steroids No LRT symptoms
Ribavirin (n ¼ 16)
Placebo (n ¼ 19)
2 5 2 4
3 8 1 5
TABLE 2. Demographic data for patients recruited to study. Data for mean and range shown
Gender Age (days) Weight (kg)
Ribavirin
Placebo
9m 12f 93?7 (15–188) 5?7 (3?1–9?3)
9m 10f 89?4 (16–266) 5?4 (1?9–9?0)
277
gender. Clinical severity at entry to the trial as assessed by arterial oxygen saturation in air and clinical score was not significantly different between the groups. Both groups had a median arterial oxygen saturation in air of 89% and the baseline clinical scores at entry were mean 5?9 (median 6?0) in the placebo group compared to the ribavarin group mean 6?0 (median 6?0). Of the 40 patients entered into the study 38 were RSV-positive. Of the two RSVnegative patients one received placebo and one ribavirin. They both had relatively long admissions of 8 and 11 days respectively. During the acute illness there was no statistical difference in the rate of improvement of clinical scores, duration of oxygen therapy or time to discharge (Table 3). The duration of hospitalization was very similar to that reported for centres in the U.K., U.S.A. and Australia (25). There were no significant differences in the frequency of reported lower respiratory tract symptoms between the two groups in the 35 patients followed to 1 year post-discharge or in any of the variables noted in Table 1. Four patients were lost to follow-up and one patient, who had received ribavirin, died some months after discharge from an illness consistent with an autoimmune enteropathy. A total of 28 patients underwent assessment of bronchial hyper-responsiveness, of whom five woke before completion of the study. A small number of patients were lost to follow-up as noted above. Lung function measurements were not obtained in others because their parents declined to have their infants sedated. Increased bronchial responsiveness defined as a fall in sGaw of 420% was noted in three of the 11 patients treated with ribavirin and in one of 12 patients treated with placebo.
Discussion This study was unable to show any significant benefit when ribavirin was used to treat previously healthy infants admitted with moderately severe bronchiolitis. This applies to both the acute illness and to subsequent respiratory morbidity. Furthermore, we were unable to show any difference in bronchial hyper-responsiveness 6 months after the acute illness. This study does not support the routine use of this drug in the treatment of acute bronchiolitis in previously healthy individuals. Other studies utilising ribavirin to treat previously healthy infants have included relatively few patients and have not provided any evidence that ribavirin has a significant impact on the clinical course of infants admitted with acute bronchiolitis. A recent systematic review (26) and meta-analysis which included studies involving high risk patients was unable to identify evidence that ribavirin has any significant impact on the course of the disease. They concluded that ribavirin does not significantly reduce mortality or duration of hospitalization, prevent deterioration or reduce duration of oxygen dependency. Our results support these conclusions, though interestingly those patients treated with ribavirin tended to remain in hospital longer than those treated with placebo. There was one patient who remained in hospital significantly longer than
278
M. L. EVERARD
ET AL.
TABLE 3. Ribavirin vs. controls: comparison of mean values both unadjusted and adjusted for age at admission, weight and sex Variable
Days in O2 Unadjusted Adjusted Change in score (day 1, day 0) Unadjusted Adjusted Change in oxygen saturation measured in air (day 17day 0) Unadjusted Adjusted Days to discharge Unadjusted Adjusted Days fit for discharge Unadjusted Adjusted
Ribavarin
Placebo
Mean difference (95% CI)
P-value
3?36 2?72
2?52 1?92
0?84 (70?63, 2?31) 0?80 (70?73, 2?32)
0?41 0?43
70?83 72?11
71?05 72?25
0?22 (70?67, 1?11) 0?14 (70?72, 1?01)
0?83 0?89
2?05 4?03
0?57 2?47
1?48 (71?59, 4?55) 1?56 (71?56, 4?68)
0?15 0?13
5?58 4?94
3?95 3?36
1?63 (70?09, 3?35) 1?58 (70?19, 3?35)
0?11 0?12
4?77 3?43
3?86 2?69
0?91 (70?63, 2?44) 0?74 (70?82, 2?30)
0?37 0?46
Notes. All P-values not statistically significant at an arbitary 5% level. Adjusted mean differences from multiple regression analyses. For all five regression analyses, the residuals followed an approximate normal distribution. Changes in score at day 1 not recorded for one infant (ribavirin). Days fit for discharge not recorded in two babies (both ribavirin).
the others and who may have aspirated during the convalescent period. However, this trend persists even when this patient was removed from the analysis. A recently published retrospective, non-randomized study with a high drop-out rate (27) suggested that patients treated with ribavirin reduced the frequency of lower respiratory tract symptom in the following year. We found no evidence for such an effect and indeed our findings are consistent with two recently published studies which followed up patients included in randomized placebocontrolled trials 10 years after their admission to hospital (28,29). These studies found no evidence that ribavirin had a long-term beneficial effect on symptoms or lung function. If ribavirin therapy were to have an effect on subsequent symptoms it is necessary to propose that the illness itself is capable of inducing changes that predispose patients to subsequent symptoms and ribavirin prevents this. The lack of proven efficacy in the acute illness would suggest that the effects of ribavirin on the inflammatory process, if any, were relatively small and therefore unlikely to have a significant effect on the long-term sequelae. It is of course possible that ribavirin might have actions other than simply limiting viral replication that influence long term outcome but there is currently no evidence for this. Our findings that 75% of the ribavirin treated group and 73?7% of the placebo group experience further lower respiratory tract symptoms in the year following admission is in line with previously published studies (30,31). Overall, only 17% of the 23 infants undergoing distilled water challenge demonstrated evidence of bronchial hyperresponsiveness, a much lower percentage than observed in
other groups of recurrently wheezy infants (20). The low incidence of bronchial hyper-reactivity observed in these infants may indicate that bronchial hyper-responsiveness is not a major contributory factor in the subsequent episodes of lower respiratory tract symptoms. As noted in the introduction, the relationship between the acute illness and subsequent recurrent respiratory morbidity remains unclear. The long running debate surrounding the use of ribavirin in acute bronchiolitis is in large part due to the fact that it was introduced with limited data to support its use. Early studies with ribavirin contained relatively few patients in stark contrast to the recent large randomized double-blind, placebo-controlled studies involving more than 1000 patients assessing the ability of synthetic monoclonal antibodies to prevent significant disease in high risk infants (32). It is possible that significant differences may have been observed in our study if larger numbers had been included. If this study were to be repeated a sample size of 104 patients would be required to show a significant difference in days to discharge using a power of 80% and a significance level of 5% though it should be remembered that this is based on a trend to earlier discharge in the placebo group. The very small differences in the other outcomes would indicate that much larger numbers than 104 would be required to demonstrate a significant difference. It seems unlikely that such small differences could translate into clinically significant effects. In summary, this prospective, randomized, placebocontrolled trial failed to identify any short- or long-term
RIBAVIRIN: ACUTE AND CHRONIC RESPIRATORY MORBIDITY benefits that could be attributed to the use of ribavirin to treat previously healthy infants.
Acknowledgements The ribavirin used in this study was provided by ICN Pharmaceuticals.
References 1. Kim HW, Arrabio JO, Brandt CD, et al. Epidemiology of respiratory syncitial virus infection in Washington DC. 1. Importance of the virus in different respiratory tract disease syndromes and temporal distribution of infection. Am J Epidemiol 1973; 98: 2l6–225. 2. Martin AJ, Gardner PS, McQuillin J. Epidemiology of respiratory viral infection among paediatric inpatients over a six year period in North East England. Lancet 1978; 2: 1035–1038. 3. Gilchrist S, Torok TJ, Gary HE, Jr, Alexander JP, Anderson LJ. National surveillance for respiratory syncytial virus, United States, 1985–1990. J Inf Dis 1994; 170: 986–990. 4. Everard ML. Respiratory syncytial virus bronchiolitis and pneumonia. In: Textbook of Pediatric Respiratory Medicine. Taussig L, Landau L, eds. St. Louis: Mosby, 1998; 580–594. 5. Rakshi K, Couriel JM. Personal practice: management of acute bronchiolitis. Arch Dis Child 1994; 71: 463–469. 6. American Academy of Pediatrics Committee on Infectious Diseases. Reassessment of the indication for ribavirin therapy in respiratory syncytial virus infections. Pediatrics 1996; 97: 137–140. 7. Pullen CR, Hey EN. Wheezing, asthma and pulmonary dysfunction 10 years after infection with respiratory syncytial virus in infancy. BMJ 1982; 284: 1665–1669. 8. Noble V, Murray M, Webb MSC, Alexander J, Swarbrick AS, Milner AD. Respiratory status and allergy nine to 10 years after acute bronchiolitis. Arch Dis Child 1992; 67: 482–487. 9. Carlsen KH, Larsen S, Orstavik I. Acute bronchiolitis. The relationship to later recurrent obstructive airways disease. Eur J Respir Dis 1987; 70: 86–92. 10. McBride JT, McConnochie KM. RSV, recurrent wheezing and ribavirin. Pediatr Pulmonol 1998; 25: 145–146. 11. Everard ML, Swarbrick A, Wrightman M, et al. Analysis of cells obtained by bronchial lavage of infants with respiratory syncytial virus infection. Arch Dis Child 1994; 71: 428–432. 12. Welliver RC, Sun M, Rinaldo D, Ogra PL. Predictive valve of respiratory syncitial virus-specific IgE responses for recurrent wheezing following bronchiolitis. J Pediatrics 1986; l09: 776–800.
279
13. Everard ML, Fox G, Walls AF, et al. Tryptase and IgE concentrations in the respiratory tract of infants with acute bronchiolitis. Arch Dis Child 1995; 72: 64–69. 14. Toms GL, Quinn R, Robinson JW. Undetectable IgE responses after respiratory syncytial virus infection. Arch Dis Child 1996; 74: 126–130. 15. Sigurs N, Bjarnason R, Sigurbergsson F, Kjellman B, Bjorksten B. Asthma and immunoglobulin E antibodies after respiratory syncytial virus bronchiolitis: a prospective cohort study with matched controls. Pediatrics 1995; 95: 500–505. 16. Everard ML. Acute bronchiolitis — a perenial problem. Lancet 1996: 348: 279–280. 17. Phelan PD, Olinsky A, Roberton CF. Respiratory Illness in Children. 4th edn. Oxford: Blackwell Scientific Publication, l994. 18. Moler FW, Bandy KP, Custer JR. Ribavirin therapy for acute bronchiolitis: neeed for appropriate controls. J Pediatr 1991; 119: 509. 19. McBride JT, Wooden KA, Katz R. Reply. J Pediatr 1991; 119: 510. 20. O’Callaghan C, Milner AD, Webb MSC, Swarbrick A. Nebulised water as a bronchoconstricting challenge in infancy. Arch Dis Child 1991; 66: 948–951. 21. Gutkowski P. Airway responsiveness following wheezy bronchitis in infants. Eur Respir J 1990; 3: 807–811. 22. Prendiville A, Green S, Silverman M. Bronchial responsiveness to histamine in wheezy infants. Thorax 1987; 42: 92–99. 23. Prendiville A, Green S, Silverman M. Airways responsiveness in wheezy infants. Evidence for functional b receptors. Thorax 1987; 42: 100–104. 24. Stocks J, Marchal F, Kraemer R, Gutkowski P, Bar Yishay E, Godfrey S. Plethysmographic assessment of functional residual capacity and airways resistance. In: Infant Respiratory Function Testing. Stocks J, Sly PD, Tepper RS Morgan WJ, eds. New York: Wiley-Liss, 1996; 191–139. 25. Behrendt CE, Decker MD, Burch DJ, Watson PH. International variation in the management of infants hospitalized with respiratory syncytial virus. Eur J Pediatr 1998; 157: 215–220. 26. Randolph AG, Wang EEL. Ribavirin for respiratory syncytial virus lower respiratory tract infection. Arch Pediatr Adolesc Med 1996; 150: 942–947. 27. Edell D, Bruce E, Hale K, Edell D, Khoshoo V. Reduced long term respiratory morbidity after treatment of respiratory syncytial virus bronchiolitis with ribavirin in previously healthy infants: a preliminary report. Pediatr Pulmonol 1998; 25: 145–146. 28. Long CE, Voter KZ, Barker WH, Hall CB. Longterm follow up of children hospitalized with respiratory syncytial virus lower respiratory tract infection and randomly treated with ribavirin or placebo. Pediatr Infect Dis J 1997: 16: 1023–1028. 29. Krilov LR, Mandel FS, Barone SR, Fagin JC. Followup of children with respiratory syncytial virus bronchiolitis in 1986 and 1987: potential effect of ribavirin on long-term pulmonary function. Pediatr Infect Dis J 1997: 16: 273–276.
280
M. L. EVERARD
ET AL.
30. Sly PD, Hibbert HE. Childhood asthma following hospitaliation with acute viral bronchiolitis in infancy. Paediatr Pulmonol 1989; 7: 153–158. 31. Henry RL, Hodges IGC, Milner AD, Stokes GM. Respiratory problems 2 years after acute bronchiolitis in infancy. Arch Dis Child 1983; 58: 713–716.
32. The IMpact–RSV Study Group. Palivizumab, a humanized respiratory sysncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high risk infants. Pediatrics 1998; 102: 531–537.
The effect of ribavirin to treat previously healthy infants admitted with acute bronchiolitis on acute and chronic respiratory morbidity M. L. EVERARD*, A. SWARBRICK*, A. S. RIGBY{ AND A. D. MILNER* *Paediatric Respiratory Unit, Queens Medical Centre, Nottingham, U.K. and {Division of Child Health, Sheffield Children’s Hospital, Sheffield, U.K. The role of ribavirin in the treatment of acute bronchiolitis is controversial. It has been suggested that the use of ribavirin may be of benefit during the acute illness and may reduce subsequent recurrent respiratory morbidity. This randomized, double-blind, placebo-controlled study was designed to determine whether ribavirin administered during the acute illness would have an influence on respiratory morbidity during both the acute illness and during the following year. Bronchial reactivity 6 months after the acute illness was also assessed. Forty previously well infants with moderately severe acute bronchiolitis were recruited during three winter epidemics. Subjects received study medication for 18 h a day. Management was otherwise unaltered. Subjects were evaluated daily by the investigator and subsequently assessed at 6 weeks, 6 months and 1 year following the acute illness. Assessment of bronchial hyper-responsiveness was assessed at 6 months of age using total body plethysmography and an established ultra-sonically nebulized distilled water challenge. A total of 40 patients (21 ribavirin, 19 placebo) were entered into the study. The two groups did not differ with respect to age, gender or clinical severity on entry to the trial. No significant differences were identified in the rate of clinical improvement over the first 24 h, the time to discharge, bronchial responsiveness at 6 months of age, frequency of significant respiratory symptoms over the first year of life and the frequency of prescribed bronchodilators and inhaled steroids during the year of follow-up. This study was unable to demonstrate any clinical benefit from the use of ribavirin in the acute illness or during subsequent follow-up for 1 year. Key words: respiratory syncytial virus; acute bronchiolitis; ribavirin; respiratory morbidity. RESPIR. MED. (2001) 95, 275–280
Introduction Acute bronchiolitis is the commonest form of lower respiratory tract infection in infancy, with 0?5–2% of all infants being hospitalized with this condition during the winter following their birth (1–3). In the majority of cases the causative organism is the respiratory syncytial virus (RSV). For the vast majority of infants admitted with this condition, careful monitoring and good supportive care remains the cornerstone of management (4,5). It has been suggested that the anti-viral drug, ribavirin, has a role in the treatment of these infants but despite the fact that ribavirin has been licensed for the treatment of infants with respiratory syncytial virus for more than a decade, its place in the treatment of RSV bronchiolitis remains controverReceived 15 September 2000 and accepted in revised form 2 January 2001. Correspondence should be addressed to: Dr Mark L. Everard, Paediatric Respiratory Unit, Sheffield Children’s Hospital, Western Bank, Sheffield S10 2TH, U.K. E-mail: [email protected]
0954-6111/01/040275+06 $35?00/0
# 2001 HARCOURT PUBLISHERS LTD
sial. This has been reflected in the position of the American Academy of Pediatrics recent reassessment of the indications for its use which resulted in the recommendation that practitioners should be ‘allowed to decide whether ribavirin therapy is appropriate by taking into account the particular clinical situation and their own preferences’ (6). Previous randomized double-blind studies have concentrated on the effect of this drug on the course of the acute respiratory illness. However, the morbidity associated with RSV bronchiolitis is not confined to the acute illness in that the majority of infants admitted to hospital with acute bronchiolitis will subsequently experience recurrent lower respiratory tract symptoms, particularly in the pre-school years (7–9). The cause of these recurrent respiratory symptoms is unclear (4,10). It may be that certain forms of pre-existing respiratory dysfunction predispose individual infants to both acute bronchiolitis and to the subsequent recurrent symptoms. Alternatively, changes in the lower airway structure or physiology caused by the intense inflammation present in the airways of infants with acute bronchiolitis (11) may contribute to the subsequent # 2001 HARCOURT PUBLISHERS LTD
276
M. L. EVERARD
ET AL.
respiratory morbidity. Other suggestions have included the suggestion that an RSV-specific IgE response contributes to subsequent morbidity (12), although recent studies suggest that this is unlikely (13,14). Previous studies have provided convincing evidence that RSV bronchiolitis does not promote expression of atopy (4,7), although a recent study suggests that this may be the case (15). If recurrent symptoms are attributable to the effects of virally-induced inflammation, then it is possible that the use of ribavirin early in the illness would limit viral replication, limit the intensity of the inflammatory process, and so reduce the frequency of post-bronchiolitic symptoms. This randomized, controlled, double-blind study was designed to determine whether administering ribavirin to previously healthy infants admitted with acute bronchiolitis would reduce the frequency of lower respiratory tract symptoms in the year following admission. In addition, we studied the effect of ribavirin therapy on the course of the acute illness and assessed bronchial responsiveness at 6 months following the acute illness using a nebulized distilled water challenge.
Methods Forty infants admitted to the paediatric wards of the Queens Medical Centre, Nottingham with moderately severe bronchiolitis were recruited during three RSV epidemics. All were previously fit and had no high risk factors for severe disease. All were entered into the study within 12 h of admission. To enter the study the definition of ‘bronchiolitis’ used in the U.K. and Australia was used (16,17), that is, patients with evidence of upper respiratory tract viral illness who develop lower respiratory tract involvement characterized by airways obstruction and widespread crepitations on auscultation. Wheeze and rhonchi may or may not be present. Patients were entered into the study before their RSV status was assessed by an immunofluorescence technique to ensure that they were entered into the study as soon after admission as possible in line with the suggestion that the early use of ribavirin is associated with the greatest benefit. These patients were randomized to receive 6 g of ribavirin in 180 ml of water by the Small Particle Aerosol Generator (SPAG). The medication was administered following a standard regime over 18 h per day. Normal saline was used for the placebo since previous studies had been criticised for using distilled water as placebo. It has been suggested that nebulized distilled water may adversely affect these infants by inducing bronchoconstriction (18). However, the output of the SPAG is so low that this seems unlikely (19). The study was approved by the local Ethics Committee and informed consent was obtained from the parents of all subjects.
ACUTE ILLNESS Subjects were evaluated daily until discharge and their clinical status was assessed by one investigator (M.L.E.), who undertook assessments in a blinded fashion. In order
to ensure that the assessment was blinded the SPAG was switched off and any crystalline ribavirin removed from tubing, bedding and patient before the patient was assessed. Patients were assigned a daily severity score. No scoring that accurately reflects the severity of the acute illness has been validated. The arbitrary scoring system chosen used a scale of 1–10; 1 representing no symptoms and 10 would be applied to a ventilated infant. Time until discharge from hospital was recorded with an additional outcome of ‘fit for discharge’ being recorded. ‘Fit for discharge’ was defined as time when patients were deemed to be medically fit for discharge. Time to discharge can be influenced by social factors and this second parameter was included to try and ensure social factors did not obscure treatment effects. Duration of supplementary oxygen were all recorded.
FOLLOW-UP Following discharge, subjects were reviewed at 6 weeks, 6 months and 1 year after discharge. Patients were provided with diaries at each visit in which they were asked to record all respiratory symptoms, medical attention for any respiratory symptoms and any medication prescribed for respiratory symptoms. The dates of onset and resolution of symptoms were also recorded. Parents were asked to record any of the following respiratory symptoms—running nose, cough, wheeze and shortness of breath. Bronchial hyper-responsiveness was assessed using an ultra-sonic distilled water challenge as previously established in our laboratory (20). A number of approaches to assessing bronchial hyper-responsiveness in infants using total plethysmography have been described using pharmacological agents (21–23). However we chose to use a ‘nonspecific’ hypo-osmolar challenge as it was felt that this may be more clinically relevant than using a mediator such as histamine. This approach has been used by a number of centres studying bronchial responsiveness in adults and older children. The use of an ultra-sonic distilled water challenge was evaluated in 100 infants prior to the commencement of this study (20). A positive response to the ultra-sonic challenge was defined as a fall in specific airways conductance (sGaw) of greater than 20% from baseline based on previous work assessing reproducibility of measurements (20). Airways conductance was assessed using absolute total body plethesmography which had been used to assess airways conductance for more than four decades (24) and was used in our unit for more than a decade prior to this study. The technique has been used extensively over recent years, providing insights into a wide variety of cardiopulmonary disease and response to therapy (24). The principles and practical considerations have been recently been reviewed in considerable detail (24). The total body plethysmograph used in this study was a purpose built unit constructed by the Department of Medical Physics, Queen’s Medical Centre, Nottingham based on a standard design (20). The volume of the plethysmograph was 65 l. Once sedated, with 100 mg kg71 of chloral hydrate, infants were placed in the plethysmograph and allowed to settle. Baseline
RIBAVIRIN: ACUTE AND CHRONIC RESPIRATORY MORBIDITY measurements were made when the system had equilibrated and there were minimal changes (less than 0?58) in the slope of the box pressure/slope readings from the face of the oscilloscope over a 5-min period. Infants then inhaled ultrasonically nebulized distilled water for 2 min administered via an Intersurgical Variosonic nebulizer, with the aerosol being delivered via the tube connecting the infants face mask to the outside of the box ensuring that the box and infant were not disturbed. Measurements of sGaw were made at 2-min intervals for a total of 12 min after the nebulization was completed.
STATISTICS The ribavirin and placebo groups were analysed by comparison of the means: 95% confidence intervals were calculated for the mean differences. The results presented in Table 3 give both unadjusted and adjusted mean differences (adjusting for age at admission, sex and weight of baby). The adjusted analyses were carried out by linear leastsquares regression method on each of five outcome measures (i.e. change in score, days in oxygen, changes in oxygen saturation measured in air, days to discharge and days fit for discharge). For each regression analysis, the residuals followed an approximate normal distribution. A nominal significance level of 5% was assumed throughout.
Results Demographic details of the patients are included in Table 2. The two groups did not differ in respect to age, weight or
TABLE 1. Lower respiratory tract symptoms and therapy required during the first year following admission to hospital with acute brionchiolitis
Admitted with LRT symptoms Bronchodilators Inhaled steroids No LRT symptoms
Ribavirin (n ¼ 16)
Placebo (n ¼ 19)
2 5 2 4
3 8 1 5
TABLE 2. Demographic data for patients recruited to study. Data for mean and range shown
Gender Age (days) Weight (kg)
Ribavirin
Placebo
9m 12f 93?7 (15–188) 5?7 (3?1–9?3)
9m 10f 89?4 (16–266) 5?4 (1?9–9?0)
277
gender. Clinical severity at entry to the trial as assessed by arterial oxygen saturation in air and clinical score was not significantly different between the groups. Both groups had a median arterial oxygen saturation in air of 89% and the baseline clinical scores at entry were mean 5?9 (median 6?0) in the placebo group compared to the ribavarin group mean 6?0 (median 6?0). Of the 40 patients entered into the study 38 were RSV-positive. Of the two RSVnegative patients one received placebo and one ribavirin. They both had relatively long admissions of 8 and 11 days respectively. During the acute illness there was no statistical difference in the rate of improvement of clinical scores, duration of oxygen therapy or time to discharge (Table 3). The duration of hospitalization was very similar to that reported for centres in the U.K., U.S.A. and Australia (25). There were no significant differences in the frequency of reported lower respiratory tract symptoms between the two groups in the 35 patients followed to 1 year post-discharge or in any of the variables noted in Table 1. Four patients were lost to follow-up and one patient, who had received ribavirin, died some months after discharge from an illness consistent with an autoimmune enteropathy. A total of 28 patients underwent assessment of bronchial hyper-responsiveness, of whom five woke before completion of the study. A small number of patients were lost to follow-up as noted above. Lung function measurements were not obtained in others because their parents declined to have their infants sedated. Increased bronchial responsiveness defined as a fall in sGaw of 420% was noted in three of the 11 patients treated with ribavirin and in one of 12 patients treated with placebo.
Discussion This study was unable to show any significant benefit when ribavirin was used to treat previously healthy infants admitted with moderately severe bronchiolitis. This applies to both the acute illness and to subsequent respiratory morbidity. Furthermore, we were unable to show any difference in bronchial hyper-responsiveness 6 months after the acute illness. This study does not support the routine use of this drug in the treatment of acute bronchiolitis in previously healthy individuals. Other studies utilising ribavirin to treat previously healthy infants have included relatively few patients and have not provided any evidence that ribavirin has a significant impact on the clinical course of infants admitted with acute bronchiolitis. A recent systematic review (26) and meta-analysis which included studies involving high risk patients was unable to identify evidence that ribavirin has any significant impact on the course of the disease. They concluded that ribavirin does not significantly reduce mortality or duration of hospitalization, prevent deterioration or reduce duration of oxygen dependency. Our results support these conclusions, though interestingly those patients treated with ribavirin tended to remain in hospital longer than those treated with placebo. There was one patient who remained in hospital significantly longer than
278
M. L. EVERARD
ET AL.
TABLE 3. Ribavirin vs. controls: comparison of mean values both unadjusted and adjusted for age at admission, weight and sex Variable
Days in O2 Unadjusted Adjusted Change in score (day 1, day 0) Unadjusted Adjusted Change in oxygen saturation measured in air (day 17day 0) Unadjusted Adjusted Days to discharge Unadjusted Adjusted Days fit for discharge Unadjusted Adjusted
Ribavarin
Placebo
Mean difference (95% CI)
P-value
3?36 2?72
2?52 1?92
0?84 (70?63, 2?31) 0?80 (70?73, 2?32)
0?41 0?43
70?83 72?11
71?05 72?25
0?22 (70?67, 1?11) 0?14 (70?72, 1?01)
0?83 0?89
2?05 4?03
0?57 2?47
1?48 (71?59, 4?55) 1?56 (71?56, 4?68)
0?15 0?13
5?58 4?94
3?95 3?36
1?63 (70?09, 3?35) 1?58 (70?19, 3?35)
0?11 0?12
4?77 3?43
3?86 2?69
0?91 (70?63, 2?44) 0?74 (70?82, 2?30)
0?37 0?46
Notes. All P-values not statistically significant at an arbitary 5% level. Adjusted mean differences from multiple regression analyses. For all five regression analyses, the residuals followed an approximate normal distribution. Changes in score at day 1 not recorded for one infant (ribavirin). Days fit for discharge not recorded in two babies (both ribavirin).
the others and who may have aspirated during the convalescent period. However, this trend persists even when this patient was removed from the analysis. A recently published retrospective, non-randomized study with a high drop-out rate (27) suggested that patients treated with ribavirin reduced the frequency of lower respiratory tract symptom in the following year. We found no evidence for such an effect and indeed our findings are consistent with two recently published studies which followed up patients included in randomized placebocontrolled trials 10 years after their admission to hospital (28,29). These studies found no evidence that ribavirin had a long-term beneficial effect on symptoms or lung function. If ribavirin therapy were to have an effect on subsequent symptoms it is necessary to propose that the illness itself is capable of inducing changes that predispose patients to subsequent symptoms and ribavirin prevents this. The lack of proven efficacy in the acute illness would suggest that the effects of ribavirin on the inflammatory process, if any, were relatively small and therefore unlikely to have a significant effect on the long-term sequelae. It is of course possible that ribavirin might have actions other than simply limiting viral replication that influence long term outcome but there is currently no evidence for this. Our findings that 75% of the ribavirin treated group and 73?7% of the placebo group experience further lower respiratory tract symptoms in the year following admission is in line with previously published studies (30,31). Overall, only 17% of the 23 infants undergoing distilled water challenge demonstrated evidence of bronchial hyperresponsiveness, a much lower percentage than observed in
other groups of recurrently wheezy infants (20). The low incidence of bronchial hyper-reactivity observed in these infants may indicate that bronchial hyper-responsiveness is not a major contributory factor in the subsequent episodes of lower respiratory tract symptoms. As noted in the introduction, the relationship between the acute illness and subsequent recurrent respiratory morbidity remains unclear. The long running debate surrounding the use of ribavirin in acute bronchiolitis is in large part due to the fact that it was introduced with limited data to support its use. Early studies with ribavirin contained relatively few patients in stark contrast to the recent large randomized double-blind, placebo-controlled studies involving more than 1000 patients assessing the ability of synthetic monoclonal antibodies to prevent significant disease in high risk infants (32). It is possible that significant differences may have been observed in our study if larger numbers had been included. If this study were to be repeated a sample size of 104 patients would be required to show a significant difference in days to discharge using a power of 80% and a significance level of 5% though it should be remembered that this is based on a trend to earlier discharge in the placebo group. The very small differences in the other outcomes would indicate that much larger numbers than 104 would be required to demonstrate a significant difference. It seems unlikely that such small differences could translate into clinically significant effects. In summary, this prospective, randomized, placebocontrolled trial failed to identify any short- or long-term
RIBAVIRIN: ACUTE AND CHRONIC RESPIRATORY MORBIDITY benefits that could be attributed to the use of ribavirin to treat previously healthy infants.
Acknowledgements The ribavirin used in this study was provided by ICN Pharmaceuticals.
References 1. Kim HW, Arrabio JO, Brandt CD, et al. Epidemiology of respiratory syncitial virus infection in Washington DC. 1. Importance of the virus in different respiratory tract disease syndromes and temporal distribution of infection. Am J Epidemiol 1973; 98: 2l6–225. 2. Martin AJ, Gardner PS, McQuillin J. Epidemiology of respiratory viral infection among paediatric inpatients over a six year period in North East England. Lancet 1978; 2: 1035–1038. 3. Gilchrist S, Torok TJ, Gary HE, Jr, Alexander JP, Anderson LJ. National surveillance for respiratory syncytial virus, United States, 1985–1990. J Inf Dis 1994; 170: 986–990. 4. Everard ML. Respiratory syncytial virus bronchiolitis and pneumonia. In: Textbook of Pediatric Respiratory Medicine. Taussig L, Landau L, eds. St. Louis: Mosby, 1998; 580–594. 5. Rakshi K, Couriel JM. Personal practice: management of acute bronchiolitis. Arch Dis Child 1994; 71: 463–469. 6. American Academy of Pediatrics Committee on Infectious Diseases. Reassessment of the indication for ribavirin therapy in respiratory syncytial virus infections. Pediatrics 1996; 97: 137–140. 7. Pullen CR, Hey EN. Wheezing, asthma and pulmonary dysfunction 10 years after infection with respiratory syncytial virus in infancy. BMJ 1982; 284: 1665–1669. 8. Noble V, Murray M, Webb MSC, Alexander J, Swarbrick AS, Milner AD. Respiratory status and allergy nine to 10 years after acute bronchiolitis. Arch Dis Child 1992; 67: 482–487. 9. Carlsen KH, Larsen S, Orstavik I. Acute bronchiolitis. The relationship to later recurrent obstructive airways disease. Eur J Respir Dis 1987; 70: 86–92. 10. McBride JT, McConnochie KM. RSV, recurrent wheezing and ribavirin. Pediatr Pulmonol 1998; 25: 145–146. 11. Everard ML, Swarbrick A, Wrightman M, et al. Analysis of cells obtained by bronchial lavage of infants with respiratory syncytial virus infection. Arch Dis Child 1994; 71: 428–432. 12. Welliver RC, Sun M, Rinaldo D, Ogra PL. Predictive valve of respiratory syncitial virus-specific IgE responses for recurrent wheezing following bronchiolitis. J Pediatrics 1986; l09: 776–800.
279
13. Everard ML, Fox G, Walls AF, et al. Tryptase and IgE concentrations in the respiratory tract of infants with acute bronchiolitis. Arch Dis Child 1995; 72: 64–69. 14. Toms GL, Quinn R, Robinson JW. Undetectable IgE responses after respiratory syncytial virus infection. Arch Dis Child 1996; 74: 126–130. 15. Sigurs N, Bjarnason R, Sigurbergsson F, Kjellman B, Bjorksten B. Asthma and immunoglobulin E antibodies after respiratory syncytial virus bronchiolitis: a prospective cohort study with matched controls. Pediatrics 1995; 95: 500–505. 16. Everard ML. Acute bronchiolitis — a perenial problem. Lancet 1996: 348: 279–280. 17. Phelan PD, Olinsky A, Roberton CF. Respiratory Illness in Children. 4th edn. Oxford: Blackwell Scientific Publication, l994. 18. Moler FW, Bandy KP, Custer JR. Ribavirin therapy for acute bronchiolitis: neeed for appropriate controls. J Pediatr 1991; 119: 509. 19. McBride JT, Wooden KA, Katz R. Reply. J Pediatr 1991; 119: 510. 20. O’Callaghan C, Milner AD, Webb MSC, Swarbrick A. Nebulised water as a bronchoconstricting challenge in infancy. Arch Dis Child 1991; 66: 948–951. 21. Gutkowski P. Airway responsiveness following wheezy bronchitis in infants. Eur Respir J 1990; 3: 807–811. 22. Prendiville A, Green S, Silverman M. Bronchial responsiveness to histamine in wheezy infants. Thorax 1987; 42: 92–99. 23. Prendiville A, Green S, Silverman M. Airways responsiveness in wheezy infants. Evidence for functional b receptors. Thorax 1987; 42: 100–104. 24. Stocks J, Marchal F, Kraemer R, Gutkowski P, Bar Yishay E, Godfrey S. Plethysmographic assessment of functional residual capacity and airways resistance. In: Infant Respiratory Function Testing. Stocks J, Sly PD, Tepper RS Morgan WJ, eds. New York: Wiley-Liss, 1996; 191–139. 25. Behrendt CE, Decker MD, Burch DJ, Watson PH. International variation in the management of infants hospitalized with respiratory syncytial virus. Eur J Pediatr 1998; 157: 215–220. 26. Randolph AG, Wang EEL. Ribavirin for respiratory syncytial virus lower respiratory tract infection. Arch Pediatr Adolesc Med 1996; 150: 942–947. 27. Edell D, Bruce E, Hale K, Edell D, Khoshoo V. Reduced long term respiratory morbidity after treatment of respiratory syncytial virus bronchiolitis with ribavirin in previously healthy infants: a preliminary report. Pediatr Pulmonol 1998; 25: 145–146. 28. Long CE, Voter KZ, Barker WH, Hall CB. Longterm follow up of children hospitalized with respiratory syncytial virus lower respiratory tract infection and randomly treated with ribavirin or placebo. Pediatr Infect Dis J 1997: 16: 1023–1028. 29. Krilov LR, Mandel FS, Barone SR, Fagin JC. Followup of children with respiratory syncytial virus bronchiolitis in 1986 and 1987: potential effect of ribavirin on long-term pulmonary function. Pediatr Infect Dis J 1997: 16: 273–276.
280
M. L. EVERARD
ET AL.
30. Sly PD, Hibbert HE. Childhood asthma following hospitaliation with acute viral bronchiolitis in infancy. Paediatr Pulmonol 1989; 7: 153–158. 31. Henry RL, Hodges IGC, Milner AD, Stokes GM. Respiratory problems 2 years after acute bronchiolitis in infancy. Arch Dis Child 1983; 58: 713–716.
32. The IMpact–RSV Study Group. Palivizumab, a humanized respiratory sysncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high risk infants. Pediatrics 1998; 102: 531–537.