Table 2.
Quality assessment | No of patients | Effect | Quality | Importance | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
No of studies | Design | Limitations | Inconsistency | Indirectness | Imprecision | Other considerations | HFNC | COT/NIV | Relative (95% CI) | Absolute | ||
Intubation rate of HFNC vs. COT as a primary mode | ||||||||||||
5 | randomised trials | seriousa | no serious inconsistency | no serious indirectness | no serious imprecision | reporting biasb | 46/434 (10.6%) | 50/397 (12.6%) | OR 0.74 (0.45 to 1.21) |
30 fewer per 1000 (from 65 fewer to 23 more) |
⊕ ⊕ ΟΟ LOW | CRITICAL |
Reintubation rate of HFNC vs. COT after extubation | ||||||||||||
8 | randomised trials | seriousa | no serious inconsistency | no serious indirectness | no serious imprecision | reporting biasb strong associationc |
63/839 (7.5%) | 123/833 (14.8%) | OR 0.47 (0.29 to 0.76) |
72 fewer per 1000 (from 31 fewer to 100 fewer) |
⊕ ⊕ ⊕Ο MODERATE | CRITICAL |
Intubation rate of HFNC vs. NIV as a primary mode | ||||||||||||
2 | randomised trials | seriousa | no serious inconsistency | no serious indirectness | no serious imprecision | reporting biasd | 47/210 (22.4%) | 68/210 (32.4%) | OR 0.57 (0.36 to 0.92) |
109 fewer per 1000 (from 18 fewer to 173 more) |
⊕ ⊕ ΟΟ LOW | CRITICAL |
Reintubation rate of HFNC vs. NIV after extubation | ||||||||||||
2 | randomised trials | seriousa | no serious inconsistency | no serious indirectness | no serious imprecision | reporting biasd | 118/704 (16.8%) | 123/730 (16.8%) | OR 1.00 (0.76 to 1.32) |
0 fewer per 1000 (from 35 fewer to 43 more) |
⊕ ⊕ ΟΟ LOW | CRITICAL |
Treatment failure of HFNC vs. COT as a primary mode | ||||||||||||
5 | randomised trials | seriousa | no serious inconsistency | no serious indirectness | no serious imprecision | reporting biasb | 58/434 (13.4%) | 71/397 (17.9%) | OR 0.65 (0.43 to 0.98) |
55 fewer per 1000 (from 3 fewer to 93 fewer) |
⊕ ⊕ ΟΟ LOW | CRITICAL |
Treatment failure of HFNC vs. COT after extubation | ||||||||||||
8 | randomised trials | seriousa | serious inconsistencye | no serious indirectness | no serious imprecision | reporting biasbstrong associationc | 108/893 (12.9%) | 192/833 (23%) | OR 0.43 (0.25 to 0.73) |
116 fewer per 1000 (from 51 fewer to 161 fewer) |
⊕ ⊕ ⊕Ο MODERATE | CRITICAL |
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
CI Confidence interval, OR Odds ratio
aLack of blinding
bFunnel plot showed potential publication bias when HFNC vs. COT
cOR < 0.5
dFunnel plot showed potential publication bias when HFNC vs. NIV
eI2 = 66%