for the main comparison.
| Heliox inhalation compared with air/oxygen for acute bronchiolitis | ||||||
|
Patient or population: infants with bronchiolitis Settings: paediatric intensive care unit and emergency department Intervention: heliox inhalation Comparison: air or oxygen inhalation | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect (95% CI) | No. of participants (studies) | Quality of the evidence (GRADE) | Comments | |
| Assumed risk | Corresponding risk | |||||
| Placebo | Heliox | |||||
| Need for mechanical ventilation (invasive or not) Follow‐up in the acute phase of bronchiolitis | 155 per 1000 | 148 per 1000 (94 to 235) | RR 0.95 (0.61 to 1.49) | 408 (4 trials) | ⊕⊕⊕⊕ high |
1, 2, 3, 4 |
| Rate of intubation (all reasons) Follow‐up in the acute phase of bronchiolitis | 19 per 1000 |
54 per 1000 (18 to 168) |
RR 2.73 (0.96 to 7.75) |
408 (4 trials) | ⊕⊕⊖⊖ low |
1, 5, 3, 6 |
| Rate of emergency department discharge Follow‐up at 240 min | 229 per 1000 | 118 per 1000 (39 to 355) |
RR 0.51 (0.17 to 1.55) |
69 (1 trial) |
⊕⊕⊕⊖ moderate |
7 |
| Length of treatment Follow‐up in the acute phase of bronchiolitis | The mean length of treatment ranged across control groups from 2.04 to 2.63 days | The mean length of treatment in the intervention groups was 0.19 days lower (0.56 lower to 0.19 higher) |
— | 320 (2 trials) |
⊕⊕⊕⊖ moderate |
1, 2, 8 |
| Length of treatment ‐ nCPAP Follow‐up in the acute phase of bronchiolitis | The mean length of treatment
ranged across the control
subgroup (with nCPAP) from 1.70 to 3.06 days |
The mean length of treatment in the intervention subgroup (with nCPAP) was 0.76 days lower (1.45 to 0.08 lower) |
— | 21 (1 trial) |
⊕⊕⊖⊖ low |
9 |
| Change in clinical respiratory scores (m‐WCAS) within the first hour after starting heliox treatment Follow‐up at the first hour after starting treatment | The mean change in m‐WCAS within the first hour after starting heliox treatment ranged across control groups from ‐0.35 to +0.04 | The mean change in m‐WCAS in the intervention groups was1.04 lower (1.60 to 0.48 lower) |
— | 138 (4 trials) | ⊕⊕⊕⊖ moderate |
10, 11, 12 |
| *Basis for the comparative risks: Theassumed risk is based on the Control group risk or the median control group risk (95% CI) across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio (Fixed effect); nCPAP: nasal continuous positive airway pressure; m‐WCAS: modified Wood Clinical Asthma Score | ||||||
| GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate. | ||||||
1The largest trial was at unclear risk of bias for blinding but at low risk for other bias. 2No serious inconsistency: statistical heterogeneity was low: I2 statistic = 0%. 3No serious indirectness: the four trials were conducted in France, the United Kingdom and the USA. 4The 95% CI around the risk ratio is narrow and excludes clinically important effects. 5Downgraded by 1 for serious inconsistency: heterogeneity is present: I2 statistic = 20%.
6Downgraded by 1 for serious imprecision: the 95% CI around the risk ratio is wide. 7Downgraded by 1 for serious risk of bias: only one trial is included. 8Downgraded by 1: the 95% CI values around the mean change are under the unit (fraction of a day) and could be imprecise. 9Downgraded by 2: only results from a subgroup of one trial are included. 10The two trials are at low risk of bias. 11Downgraded by 1: statistical heterogeneity was high (I2 statistic = 80%) and remained when we kept only low risk of bias trials. 12No serious indirectness: the four trials were conducted in France, the USA, the United Kingdom and Spain.