Summary of findings 1. High‐dose compared to standard‐dose strategies for any indication for preterm infants with or at risk for apnea of prematurity.
| High‐dose compared to standard‐dose strategies for any indication for preterm infants with or at risk for apnea of prematurity | ||||||
| Patient or population: preterm infants with or at risk for apnea of prematurity Setting: neonatal intensive care units Intervention: high‐dose strategies Comparison: standard‐dose strategies | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with standard‐dose strategies | Risk with high‐dose strategies | |||||
| All‐cause mortality prior to hospital discharge | Study population | RR 0.86
(0.53 to 1.38) RD ‐0.01 (‐0.05 to 0.03) |
723 (5 RCTs) | ⊕⊕⊝⊝ Lowa,b | High‐dose caffeine may have little or no effect in reducing all‐cause mortality prior to hospital discharge compared with standard‐dose caffeine. | |
| 91 per 1000 | 78 per 1000 (48 to 125) | |||||
| Major neurodevelopmental disability in children aged 18 to 24 months CA | See comments | ‐ | ‐ | ‐ | ‐ | This outcome was not reported. |
| Major neurodevelopmental disability in children aged 3 to 5 years CA | Study population | RR 0.79 (0.51 to 1.24 RD ‐0.15 (‐0.42 to 0.13) |
46 (1 RCT) | ⊕⊝⊝⊝ Very lowc,d | We are uncertain whether high‐dose caffeine improves major neurodevelopmental disability in children aged three to five years compared with standard‐dose caffeine. | |
| 720 per 1000 | ||||||
| Mortality or major neurodevelopmental disability in children aged 18 to 24 months and 3 to 5 years CA | See comments | ‐ | ‐ | ‐ | ‐ | This outcome was not reported. |
| Bronchopulmonary dysplasia/chronic lung disease (BPD) at 36 weeks' postmenstrual age | Study population | RR 0.75
(0.60 to 0.94) RD ‐0.08 (‐0.15 to ‐0.02) NNTB = 13 |
723 (5 RCTs) | ⊕⊕⊕⊝ Moderatea | High‐dose caffeine probably reduces bronchopulmonary dysplasia at 36 weeks' postmenstrual age compared with standard‐dose caffeine. | |
| 335 per 1000 | 251 per 1000 (201 to 315) | |||||
| Side effects (tachycardia, agitation, or feed intolerance) leading to a reduction in dose or withholding of caffeine | Study population | RR 1.66
(0.86 to 3.23) RD 0.03 (‐0.01 to 0.07) |
593 (5 RCTs) | ⊕⊕⊝⊝ Lowb,e | High‐dose caffeine may have little or no effect in reducing side effects (tachycardia, agitation, or feed intolerance) compared with standard‐dose caffeine. | |
| 43 per 1000 | 72 per 1000 (37 to 139) | |||||
| Duration of hospital stay (days) | See comments | ‐ | (3 RCTs) | ‐ | Three studies reported on this outcome (Mohammed 2015; Mohd 2021; Zhao 2016). Data could not be pooled in a meta‐analysis because outcomes were expressed as medians and IQRs. Mohammed 2015: high dose (median 30.5 [IQR 20 to 51.5] days), standard dose (median 35 [IQR 25 to 51.5] days). Mohd 2021: high dose (median 52 days), standard dose (median 49.50 days). IQR not reported. Zhao 2016: high dose (median 33 [IQR 25 to 49] days), standard dose (median 39 [IQR 27 to 54] days). |
|
| Seizures (clinically diagnosed; diagnosed by electroencephalography) | Study population | RR 1.42
(0.79 to 2.53) RD 0.14 (‐0.09 to 0.36) |
74 (1 RCT) | ⊕⊝⊝⊝ Very lowd,f | We are uncertain whether high‐dose caffeine reduces seizures compared with standard‐dose caffeine. | |
| 324 per 1000 | 461 per 1000 (256 to 821) | |||||
| *The risk in the intervention group (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). CA: corrected age; CI: confidence interval; NNTB: number needed to treat to benefit; RD: risk difference; RR: risk ratio | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
aDowngraded for study limitations by one level: unclear risk of bias in four domains (selection, detection, reporting, and other bias) bDowngraded for imprecision by one level: few events and wide confidence intervals. cDowngraded for study limitations by one level: unclear risk of selection and reporting bias. dDowngraded for imprecision by two levels: one small trial with few events. eDowngraded for study limitations by one level: unclear risk of bias in four domains (selection, detection, reporting, and other bias) and high risk of performance bias. fDowngraded for study limitations by one level: unclear risk of bias in two domains (selection and other bias).