Description of studies included in review
| Study | Participants | Intervention and comparison groups | Treatment schedule and rescue medicines | Outcome measures | Study design and notes |
|---|---|---|---|---|---|
| Quinn et al41 | 95 infants (≤34 weeks GA) ventilated for hyaline membrane disease, struggling with respirator, postnatal age >4 h and <48 h | Morphine (n=29): 50 μg/kg/h infusion, increased to 100 μg/kg/h if still struggling at 2 h. Pancuronium (n=28): 100 μg/kg bolus as required to inhibit spontaneous breathing. Morphine+pancuronium (n=38): morphine 50 μg/kg/h infusion+pancuronium 100 μg/kg as required. | Treatment continued until FiO2 <0.45. Infants in morphine group allowed pancuronium if still fighting ventilator at 4 h. Infants in pancuronium group allowed morphine for painful procedures. | Plasma catecholamine levels. Blood pressure. Heart rate. Peak inspiratory pressure. FiO2. Days on ventilator. Air leaks. Incidence of IVH and PDA. Death. | Randomisation by sealed envelopes. Allocation concealment inadequate. Blinding not guaranteed. |
| Quinn et al42 | 41 infants (<34 weeks GA) ventilated for hyaline membrane disease, treated with surfactant (Curosurf) and indwelling arterial line | Morphine (n=21): loading dose 100 μg/kg/h for 2 h+25 μg/kg/h infusion. Placebo (n=20): 5% dextrose. | Treatment continued as long as infant on ventilator. Muscle relaxants allowed. | Plasma catecholamines. Blood pressure, heart rate, peak inspiratory pressure and oxygen concentration. Arterial:alveolar oxygen ratio. Behavioural pain score. Days on ventilator. Air leaks. Incidence of IVH, PDA, PVL, pneumothorax. Death before age 6 months. | Randomisation by stratifi ed table. Allocation concealment adequate. Carers and assessors stated to be blinded, except for caring doctor. |
| Pokela39 | 84 infants (term and preterm) <1 week postnatal age, ventilated for respiratory distress, with hypoxaemia and needing sedation or analgesia | Meperidine (n=42): 1 mg/kg intravenously over 1 min, 15 min before tracheal suction or routine procedures. Placebo (n=42): normal saline. | 2-h study period | Duration of hypoxaemia during procedures (primary outcome). Heart rate, tcPO2, SaO2 and mean arterial blood pressure. Novel behavioural pain scale. Plasma β-endorphin, cortisol, glucose. | Randomisation by randomisation table.* Allocation concealment adequate. Carers and assessors blinded to treatment. |
| Dyke et al34 | 26 infants (29–36 weeks GA) ventilated for hyaline membrane disease on fi rst day after birth | Morphine (n=12): loading dose 100 μg/kg over 30 min+10 μg/kg/h infusion. Placebo (n=14): 5% dextrose. | Infusion continued until weaning from intermittent mandatory ventilation or for 48 h. Pancuronium allowed if infants did not stabilise. | Primary: heart rate, blood pressure, respiratory rate; severity of respiratory distress; interaction of infant with positive pressure ventilation. Secondary: days of oxygen treatment, ventilation and hospitalisation; incidence of BPD, PVL, IVH, pneumothorax. | Randomisation by computer-generated list. Allocation concealment adequate. Carers and assessors blinded to treatment. |
| Saarenmaa et al43 | 10 infants (≥24 weeks GA), intubated, with indwelling arterial line | Alfentanil: 10 μg/kg. Alfentanil: 20 μg/kg. Placebo: normal saline. | Dosing in random order 2 min before three separate endotracheal suctions, at least 6 h apart | Behavioural pain scores. Heart rate. Arterial blood pressure and SaO2. Plasma epinephrine, norepinephrine, β-endorphin. | Randomisation by sealed envelopes. Allocation concealment not specifi ed. Double blind. Crossover trial (three arms). |
| Orsini et al38 | 20 infants (26–36 weeks GA) >1000 g birth weight, ventilated for respiratory distress syndrome, with indwelling arterial catheter | Fentanyl (n=11): loading dose 5 μg/kg over 20 min+infusion 2 μg/kg/h for 72 h, 1 μg/kg/h for next 24 h and 0.5 μg/kg/h for fi nal 24 h. Placebo (n=9): 5% dextrose. | Total, 5 days | Primary: vital signs; cortisol and 11-deoxycortisol levels; 3-methyl histidine:creatinine molar ratio; urea excretion; incidence of PDA, BPD, sepsis. Other:* behavioural pain score, incidence of CLD and IVH. | Randomisation by random number generation. Allocation concealment adequate. Carers and assessors blinded to treatment. |
| Guinsburg et al35 | 22 infants (≤32 weeks GA) ventilated since birth, postnatal age 12–48 h, indwelling arterial umbilical line | Fentanyl (n=11): 3 μg/kg. Placebo (n=11): normal saline. | Single dose over 2 min | Heart rate. Blood pressure. Arterial blood gases. Ventilator parameters (FiO2, mean airways pressure, ventilatory rate). Behavioural pain scores. Serum cortisol, growth hormone, glucose, lactate. | Randomisation by sealed envelopes. Allocation concealment not specifi ed. Assessors stated to be blinded to treatment. |
| Lago et al36 | 55 infants (26–34 weeks GA) ventilated for hyaline membrane disease, with indwelling_catheter | Fentanyl (n=26): 0.5–2 μg/kg/h infusion, adjusted to render neonate sedated but arousable. Control (n=27): no intervention. | Treatment continued as long as infant on ventilator | Primary: behavioural sedation score; severity of hyaline membrane disease; need for surfactant; evidence of clinically signifi cant PDA. Secondary: molar ratio of urine metapinephrine:nor metapinephrine; days on ventilator; days of oxygen treatment; air leak; incidence of IVH, PVL, BPD; days to exclusive enteral feeding and to reach birth weight; length of hospitalisation. | Randomisation by sealed envelopes.* Assessors stated to be blinded to treatment, but blinding unlikely. Carers not blinded. Allocation concealment not specifi ed. |
| Anand et al28 | 67 infants (24–32 weeks GA), postnatal age ≤72 h, ventilated for <8 h | Morphine (n=24): loading dose 100, 200 or 300 μg/kg+10, 20 or 30 μg/kg/h infusion for infants of 24–26, 27–29 and 30–32 weeks GA, respectively. Midazolam (n=22): loading dose 200 μg/kg+20, 40 or 60 μg/kg/h infusion for infants of 24–26, 27–29 and 30–32 weeks GA, respectively. Placebo (n=21): 5% dextrose. | Treatment continued for as long as necessary, up to 14 days. Additional analgesia with morphine bolus allowed. | Primary: incidence of adverse neurological events (neonatal death, grade III or IV IVH, PVL). Secondary: level of sedation and pain response to tracheal suctioning; incidence of pneumothorax; days of ventilation, continuous positive airway pressure and oxygen support; length of stay in ICU and hospital; neurodevelopmental outcome. | Randomisation by automated procedure in blocks, stratifi ed by centre. Allocation concealment adequate. Carers and assessors blinded to treatments. |
| Lago et al37 | 31 infants (28–36 weeks GA) ventilated for hyaline membrane disease | Fentanyl (n=15): infusion 1.5 μg/kg/h scaled down by 0.5 μg/kg/h every 24 h. Placebo (n=16): 5% dextrose. | Total 72 h treatment | Ventilator parameters. Sedation score. Severity of respiratory disorder. Radiological score. Duration of ventilation. Need for surfactant treatment. Duration of oxygen dependence. EMG activity of intercostal muscles. | Randomisation by sealed envelopes.* Assessors stated blinded to treatment; carers unaware of treatment. Allocation concealment not specifi ed. |
| Siwiec et al47 | 20 infants (26–35 weeks), birth weight 810–2750 g, ventilated for illness | Morphine (n=10): loading dose 100 μg/kg over 30 min+20 μg/kg/h infusion for 1–5 days. Control (n=10): no intervention. | Continuous infusion for 1–5 days | Pain scores.* Ventilatory parameters: mean airway pressure, ventilatory rate, FiO2. Incidence of pneumothorax, grade IV IVH, PVL, BPD. | Randomisation by sealed envelopes.* Allocation concealment not specifi ed. Carers and assessors not blinded.* |
| Simons et al46 | 150 neonates of postnatal age <3 days, ventilated <8 h | Morphine (n=73): loading dose 100 μg/kg followed by 10 μg/kg/h infusion. Placebo (n=77): 5% dextrose. | Masked treatment for ≤7 days (based on clinical conditions); after 7 days, medication weaned, stopped, or replaced by open-label morphine infusion. Additional morphine allowed if patients from either group were judged by attending doctor to be in pain or distress. | Primary: pain scores. Secondary: incidence of IVH; poor neurological outcome (grade III or IV IVH, PVL or death). Other: duration of ventilation; length of ICU stay; incidence of comorbidity (CLD, sepsis, NEC, PDA); number of painful procedures. | Randomisation by computer-generated list to select 10 random permuted blocks stratifi ed in 5 groups by GA. Allocation concealment adequate. Carers and assessors blinded. |
| Anand et al29 | 898 infants (23–32 weeks GA) intubated within 72 h of birth and ventilated for <8 h at enrollment | Morphine (n=449): loading dose 100 μg/kg followed by 10, 20 or 30 μg/kg/h infusion for infants of 23–26, 27–29, or 30–32 weeks GA, respectively. Placebo (n=449): 5% dextrose. | Additional analgesia with morphine bolus was allowed | Primary: adverse neurological events (grade III or IV IVH, PVL or death); cranial ultrasonography. Secondary:* pain score; days of ventilatory support and oxygen supplementation; days to full enteral feeding. | Randomisation by automated procedure, stratifi ed by centre and GA. Allocation concealment adequate. Carers and assessors blinded to treatment. |
↵* Obtained by personal communication with authors.
BPD, bronchopulmonary dysplasia; CLD, chronic lung disease; EMG, electromyography; FiO2, fractional inspired O2 concentration; GA, gestational age; ICU, intensive care unit; IVH, intraventricular haemorrhage; NEC, necrotising enterocolitis; PDA, patent ductus arteriosus; PVL, periventricular leukomalacia; SaO2, arterial oxygen saturation; tcPO2, transcutaneous partial pressure of oxygen.