Table 1.

Summary characteristics of included systematic reviews

First author, year	Studies	Study designs	Patients	Conditions	Interventions	Comparators	Outcomes	Conclusions
Chen (2018) [12]	13	RCTs	1515	Premature infants with apnea	High MD of caffeine citrate (10–20 mg/kg daily)	Lower caffeine citrate MD (5–10 mg/kg daily)	Rate of effective treatment, defined as successful evacuation within 72 h after treatment onset, fewer than 3 apnea episodes per day, and no significant abnormalities in respiratory rhythm. Secondary events included tachycardia, electrolyte disturbance, hypertension, hyperglycemia, feed intolerance, restlessness, and in-hospital mortality	Higher MDs of caffeine citrate appear more effective and safer than low MDs for treatment of premature apnea, despite a higher incidence of tachycardia
Henderson-Smart (2010) [14]	5	RCTs	108	Preterm neonates (born before 34 weeks gestation) requiring treatment for recurrent apnea of prematurity	Caffeine citrate 20 mg/kg LD, 5 mg/kg/24 h MD	Theophylline 6 mg/kg	Apnea failed treatment defined as no clinically important reduction in apnea (> 50% reduction), use of IPPV or death during study, mean rates of apnea, and use of IPPV. Secondary events included tachycardia or feed intolerance leading to an alteration in treatment, longer-term growth and development	There were no differences in treatment failure rate and mean apnea rate between caffeine and theophylline groups after 1–3 days treatment and 5–7 days treatment. None of the trials reported effects on growth and development
Park (2015) [13]	5	Retrospective cohort (n = 4), RCT (n = 1)	59,136	VLBW infants (birth weight < 1500 g)	Early caffeine group	Late caffeine group	Death, BPD, and BPD or death. Secondary events included IVH, PVL, ROP requiring laser photocoagulation, PDA requiring treatment, NEC (medical or surgical), NEC requiring surgical treatment, and duration of mechanical ventilation	The risk of death, BPD, and BPD or death was lower in the early caffeine group. Early caffeine use was associated with reduced adverse events, but not with the risk of NEC and NEC requiring surgery
Vliegenthart (2018) [17]	6	RCTs	620	Infants with a gestational age < 32 weeks with apnea	Higher dose caffeine (10–80 LD caffeine citrate mg/kg; 5–30 MD caffeine citrate mg/kg/day)	Standard dose caffeine (10–30 LD caffeine citrate mg/kg; 2.5–10 MD caffeine citrate mg/kg/day)	Combined BPD and mortality at 36 weeks PMA, BPD at 36 weeks PMA, and death before discharge. Secondary events included tachycardia, failure to extubate, NEC ≥ grade 2, IVH ≥ grade 3 and death before 1 year CA, death or disability at 12 months CA, major disability at 12 months CA, GQ at 12 months CA, mortality < 12 months CA	There was a significant decrease in BPD, the combined outcome BPD or mortality, and failure to extubate in infants allocated to a higher caffeine dose. However, no differences were found in mortality alone and NDI
Henderson-Smart (2000) [15]	4	RCTs	91	Preterm infants with recurrent apnea	Doxapram 1.5–3.0 mg/kg/h IV	Methylxanthine 6–8 mg/kg LD, and 1.5 mg/kg/8 h MD	Failed treatment (no clinically important 50% reduction in apnea or use of IPPV or death during study) and use of IPPV. Secondary events included tachycardia, seizures or hypertension, and long-term growth and development	There was no difference found between IV doxapram and methylxanthine in the incidence of failed treatment within 48 h. None of the studies reported adverse effects
Henderson-Smart (2010) [16]	6	RCTs	959	Preterm infants with recurrent apnea	Methylxanthine (aminophylline, theophylline, caffeine)	Placebo or no treatment	Failed treatment (less than 50% reduction in apnea, or use of IPPV, or death during study, use of IPPV, and death before hospital discharge. Secondary events included tachycardia or feed intolerance leading to omission of treatment, neonatal morbidity such as PDA requiring treatment, intracranial hemorrhage, NEC, duration of IPPV and oxygen therapy, chronic lung disease indicated by respiratory support (oxygen and/or positive airway pressure)	Methylxanthine therapy led to a reduction in apnea and the use of IPPV in the first 2–7 days
Brattström (2019) [18]	6	RCTs	816	Preterm infants born before gestational week 34 admitted to neonatal intensive care units	High-dose caffeine citrate (> 20 mg/kg body weight/day and maintenance dosage > 10 mg/kg/day)	Low-dose caffeine citrate at loading dosage equal or less than 20 mg/kg/day and maintenance dosage equal or less than 10 mg of caffeine citrate/kg/day	Mortality during the first admission, BPD at 36 weeks of CA 23 and cerebral palsy. Secondary events included neonatal mortality, IVH ≥ grade 3; IVH any grade, cerebellar hemorrhage, PVL; and lesions indicative of brain injury (any lesion detected by ultrasound or MRI)	Based on limited evidence, a conclusion could not be made regarding the safety of high- versus low-dose caffeine

BPD bronchopulmonary dysplasia, CA corrected age, GQ general quotient, IPPV intermittent positive pressure ventilation, IV intravenous, IVH intraventricular hemorrhage, LD loading dose, MRI magnetic resonance imaging, MD maintenance dose, NDI neurodevelopmental impairment, NEC necrotizing enterocolitis, PDA patent ductus arteriosus, PMA post-menstrual age, PVL periventricular leukomalacia, RCT randomized controlled trials, ROP retinopathy of prematurity, VLBW very low birth weight