Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Ribavirin for Respiratory Syncytial Virus Lower Respiratory Tract Infection: A Systematic Overview

Randolph AG, Wang EE.

Arch Pediatr Adolesc Med 1996;150:942-7. [abstract]

Reviewed by Michael J. Verive, MD, Hope Children's Hospital, Oak Lawn, IL

Review posted April 19, 1999

---

I. Are the results of the study valid?

A. Primary questions:

1. Did the overview address a focused clinical question?

Yes. The overview was designed "to systematically review evidence evaluating the efficacy of aerosolized ribavirin for the treatment of infants with respiratory syncytial virus lower respiratory tract infection." What is not initially clear is what specific outcomes would be considered, but this question is dealt with in detail later in the review.

2. Were the criteria used to select articles for inclusion appropriate?

Yes. The overview specifically included only randomized, placebo-controlled studies of infants with documented RSV lower respiratory tract infection.

B. Secondary questions:

3. Is it unlikely that important, relevant studies were missed?

Yes. A computerized search of MEDLINE from 1975 to 1996 was performed, along with references from each retrieved article, and contact with experts. MEDLINE was searched using "ribavirin" as subject and "respiratory syncytial virus" as keyword, resulting in 133 citations retrieved, including 8 review articles. Forty-two additional studies were identified using references listed for the review articles, and 12 randomized clinical trials were located.

Eight published studies met inclusion criteria. Four randomized controlled trials were excluded (1 without placebo control, 1 randomized only half of the patients, 1 included patients intubated for RSV-associated apnea, 1 did not assess outcomes in manner comparable to other studies and included RSV-negative patients).

4. Was the validity of the included studies appraised?

Yes. A 14-point scale of criteria for assessment of methodologic quality was used to evaluate each study reviewed. This scale included criteria for patient selection and characteristics, randomization, control for confounding variables, and outcome assessment. Each study was ranked using the above criteria, with the ranking reported in the overview. Selected studies received quality scores ranging from 5 to 13.

5. Were assessments of studies reproducible?

Yes. Two independent observers evaluated the studies, with an appropriate kappa statistic (0.78 to 0.91 for each study across quality items) revealing sufficient inter-observer agreement in assessment.

6. Were the results similar from study to study?

Not all studies evaluated the same outcomes, but for those studies with similar outcomes a test of homogeneity (Breslow-Day test) was performed to ensure that the study outcomes were similar from study to study. Only 3 of 8 selected studies could be combined to evaluate the outcomes of mortality and respiratory deterioration. Breslow-Day test for homogeneity was non-significant for both of these. The non-significant test of homogeneity suggests that any difference in outcomes between the selected studies is likely due to chance rather than differences in study design.

II. What are the results?

1. What are the overall results of the review?

There was a trend towards a decreased mortality rate and a decreased rate of respiratory deterioration with ribavirin across the studies. The relative risk of mortality was from 0.33 to 0.5 (n=3) and the relative risk of respiratory deterioration was from 0.2 to 0.54 (n=3), respectively, with 95% confidence intervals including 1 (no effect) in all included studies. The pooled relative risk of mortality was 0.42 (95% CI=0.13, 1.44) and the pooled relative risk of respiratory deterioration was 0.42(95% CI=0.16, 1.34).

In the two studies of mechanically ventilated patients, improvements in duration of ventilation and oxygen dependence were split with one study reporting improvement in duration of ventilation and oxygen dependence in ribavirin-treated patients of -5 and -4.8 days, respectively, with 95% confidence intervals not including 0 (significant effect). The other study failed to show similar treatment effect, with improvement in duration of ventilation and oxygen dependence in ribavirin-treated patients of -1.8 and -1.4 days, respectively, with 95% confidence intervals -7.4 to 3.8 and -7.9 to 5.1 (no significant effect). The effect of ribavirin on length of hospitalization in these two studies favored ribavirin by 1.7 and 3.3 days, with 95% confidence intervals from -9.3 to 5.9 and -11.1 to 4.5, with both ranges including 0 (no effect).

Five out of six studies in non-ventilated patients assessed illness-severity scores and found significant improvement in ribavirin treated patients over time compared with controls. However, the effect of ribavirin on illness-severity scores was dependent on the subjectivity of the scoring systems used. Four of the 5 studies evaluating illness-severity scores in non-ventilated patients used a single rater, and one study used two raters but the inter-observer variability was not assessed.

Four out of six studies in non-ventilated patients assessed oxygenation change and found differences in improvement of between 1.5% and 4% (ribavirin over placebo), as measured by pulse oximetry once daily. Arterial blood gases, measured in selected patients, demonstrated improvement in PaO2 of 13 to 21.5 mm Hg for the ribavirin group vs. 4 to 1.4 mm Hg for the placebo group. The only comment that these results were significant comes in the discussion section of the paper. From five studies, no significant differences in respiratory rate, wheezing, fever, or upper respiratory tract signs were noted. In addition, there was no statistically significant difference in hepatic, renal, or hematopoietic function (3 studies).

2. How precise were the results?

Even after taking into account the variable use of saline and distilled water placebo, the results by meta-analysis revealed no statistically significant difference in mortality or respiratory deterioration. Because of differences in outcome measurements, the authors were unable to pool other measures, and only report a critical appraisal of them as noted above.

With confidence intervals crossing one for relative risk estimates and zero for duration estimates (see II.1 above), the results do not achieve statistical significance.

III. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

Possibly. Although the direction of treatment benefit from ribavirin favored treatment in each case, the lack of consistent, statistically significant improvement does not support the use of ribavirin in my patient care.

2. Were all clinically important outcomes considered?

The most clinically significant outcomes were considered by these studies, although several studies only measured a small subset of the outcomes reported.

3. Are the benefits worth the harms and costs? The studies did not address risk/benefit ratios or perform cost analyses, but considering the statistically insignificant treatment effect in the majority of the studies and the high cost of ribavirin, there is insufficient evidence to support the use of ribavirin in most patients with lower respiratory tract infection due to RSV.

Comments

Back to the EB Journal Club Index