RIBAVARIN IN VENTILATED RESPIRATORY SYNCYTIAL VIRUS BRONCHIOLITIS: A RANDOMIZED, PLACEBO-CONTROLLED TRIAL

To the Editor :

In a nicely written paper with thorough methodology, Guerguerian and colleagues (Am. J. Respir. Crit. Care Med. 1999; 160:829-834) describe 41 previously healthy infants with severe RSV bronchiolitis who had a rapidly deteriorating clinical course necessitating mechanical ventilation. No statistical difference was observed in any outcome variable between the ribavirin-treated group and controls. We would like to raise a few issues:

1. The timing of ribavirin intervention may be a crucial factor especially since ribavirin is virustatic and early intervention may produce the most benefit. It would be interesting to compare the difference in the outcome variables after initially stratifying the groups by duration of illness; i.e., early ( 4 d) and late (> 4 d).
2. The amount and distribution of ribavirin delivered in infants with severe disease requiring mechanical ventilation is always an issue. For this reason it may well be that infants with RSV requiring mechanical ventilation may not be the best candidates for ribavirin therapy.
3. The mean of all outcome variables was consistently 15- 25% better in the ribavirin group; however, the results could not achieve statistical significance because the sample size was calculated to detect a 50% difference in the outcome variables. If the sample size was much larger and if the results followed the same trend, then "magically" the outcome variables would achieve statistical significance. In real life more than 90,000 infants require hospitalization every year for RSV bronchiolitis. The real question is "What is the least acceptable improvement that will lead to a change in clinical practice?" The sample size should be calculated based on the answer to this question. Any difference smaller than that would then be regarded as being clinically not relevant.
4. On the basis of the results of this study, a mean saving of about $3,000 per patient would be expected in those treated with ribavirin, besides, of course, the potential for long-term savings.
5. We feel that oral feeding should be controlled as a confounding variable because infants with RSV bronchiolitis may be at increased risk of aspiration; thus, mild bronchiolitis with aspiration could mimic severe bronchiolitis. This could lead to an erroneous assessment of the severity of disease as well as of the duration of symptoms from RSV bronchiolitis. In the context of this study, aspiration needs to be ruled out as a cause of rapid deterioration.

West Jefferson Medical Center, New Orleans, Louisiana

From the Authors:

Although it has not been demonstrated, the timing of ribavirin administration is commonly believed to be important. Early intervention is thought to have a higher probability of influencing the course of disease. We designed our study (1) in order to administer the aerosol to a homogeneous population of infants as soon as possible after the initiation of mechanical ventilation and parental consent. Drs. Khoshoo and Edell suggest an initial stratification of the patients by duration of illness in order to determine whether early treatment has more impact on disease. However, by separating the infants into those requiring intubation more rapidly and those requiring intubation only after 4 d of symptoms, other confounding factors arise that may have an impact on the potential response to ribavirin therapy over and above the theoretical differences in their viral loads. For example, would those intubated earlier in the course of their illness, be younger, smaller, or more severely ill than the others would? Sub-group analysis after randomization would not be reliable because of the resulting smaller size in each sample.

The optimal mode of delivery of ribavirin for RSV bronchiolitis (aerosolized, enteral, parenteral) has not been studied systematically in humans. Drs. Khoshoo and Edell argue that infants with severe bronchiolitis may not be the best candidates for optimal distribution of the aerosol. However, severely ill infants and this mode of delivery are specified in standard of care guidelines such as the American Academy of Pediatrics (2) and Canadian Paediatric Society (3) recommendations. Our goal was to verify the evidence underlying these recommendations in the particular case of infants with RSV bronchiolitis needing mechanical ventilation. Other methods of delivery for this population would need to be the focus of phase I (safety) studies prior to effectiveness studies.

Sample size was calculated to enroll the necessary amount of infants to demonstrate a clinically significant outcome. In a previous published study performed in our unit (4), the mean length of mechanical ventilation was 4.4 d. We estimated that a 50% reduction in length of ventilation would be needed to justify the use of this drug and that a smaller difference would be clinically irrelevant. As stated in our conclusion, if the same trends persisted as were present in our measured outcomes and we assume them to be clinically applicable, then a sample of 290 patients would be needed to reach statistical significance. The morbidity of hospitalized infants with less severe disease is an issue that should probably be tackled with different strategies.

A cost analysis of this drug was not done as part of this study. A systematic assessment of direct and indirect costs that extend beyond the price of the compound itself would have to be generated if this drug were found to be significantly effective. Moreover, long-term savings resulting from its utilization have yet to be demonstrated.

Lastly, Drs. Kohshoo and Edell suggest that aspiration may be associated with bronchiolitis and may lead to an inaccurate assessment of severity of disease, and therefore should be ruled out. In our study, patients suffering from gastroesophageal reflux prior to their bronchiolitis episode (diagnosed either by formal diagnostic test or by their response to empiric therapy) were not eligible candidates. Moreover, we believe that documenting gastroesophageal reflux with a swallow study during an intercurrent respiratory tract infection, associated to respiratory failure, may be hazardous to the patient without demonstrating causality. The rare cases in which aspiration, occurring as a result of the acute respiratory illness, led to respiratory failure and intubation would likely have arisen to the same extent in the two groups.

ANNE-MARIE GUERGUERIAN

Pediatric Intensive Care Unit

  and Infectious Diseases

Sainte-Justine Hospital

Montréal, Québec, Canada

1. Guerguerian, A.-M., M. Gauthier, M. H. Lebel, C. A. Farell, and J. Lacroix. 1999. Ribavirin in ventilated respiratory syncytial virus bronchiolitis: a randomized, placebo-controlled trial. Am. J. Respir. Crit. Care Med. 160: 829-834 [Abstract/Free Full Text].

2. American Academy of Pediatrics, Committee on Infectious Disease. 1996. Reassessment of the indications for ribavirin therapy in respiratory syncytial virus infections. Pediatrics 97: 137-140 [Abstract/Free Full Text].

3. Canadian Paediatric Society. 1997. Use of ribavirin in the treatment of infections caused by respiratory syncytial virus. Pediatr. Child Health 2: 345-348 .

4. Lebel, M. H., M. Gauthier, J. Lacroix, E. Rousseau, and M. Buithien. 1989. Respiratory failure and mechanical ventilation in severe bronchiolitis. Arch. Dis. Child. 64: 1431-1437 [Abstract].