Table 2.
Summary of the findings
| Outcomes | Anticipated absolute effects (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | |
|---|---|---|---|---|---|
| Risk with control | Risk with melatonin | ||||
| Incidence of EA | |||||
| Melatonin and its analogs vs. placebos | 471 per 1000 |
240 fewer per 1000 (from 349 fewer to 47 fewer) |
0.49 (0.26 to 0.90) |
307 (5) |
⨁◯◯◯ Very low abc |
| Melatonin vs. placebos | 430 per 1000 | 258 fewer per 1000 (from 318 fewer to 168 fewer) |
0.40 (0.26 to 0.61) |
259 (4) |
⨁⨁⨁◯ Moderate c |
| Melatonin vs. midazolam | 311 per 1000 |
162 fewer per 1000 (from 212 fewer to 84 fewer) |
0.48 (0.32 to 0.73) |
346 (4) |
⨁⨁⨁◯ Moderate b |
| Melatonin vs. dexmedetomidine | 100 per 1000 |
104 more per 1000 (from 11 more to 273 more) |
2.04 (1.11 to 3.73) |
240 (3) |
⨁⨁⨁◯ Moderate b |
a. For high I2 scores, the level of certainty was downgraded to “serious” for “inconsistency.”
b. For the wide range of TSA-adjusted 95% CI, the certainty of evidence was downgraded to “serious” for “imprecision.”
c. Publication bias was strongly suspected
Abbreviations: EA, Emergence agitation; CI, Confidence interval; GRADE, Grading of Recommendations Assessment, Development, and Evaluation