| Methods |
Parallel randomised controlled trial |
| Participants |
1529 preterm neonates: birth weight ≤ 1500 g. Ineligible if enterally fed for > 24 hours prior to enrolment. Eligible if breast fed |
| Interventions |
Treatment commenced when enteral feeds commenced
Intervention group: oral IgG 1200 mg/kg/day only divided into at least 4 doses for 28 days. 768 neonates
Control group: placebo of oral albumin 2400 mg/kg/day, coloured to look identical to treatment. Albumin was used to simulate the viscosity of the IgG solution. 31% of neonates did not receive the study solution until after the 5th day of life due to a delay in tolerating oral feeds. 761 neonates |
| Outcomes |
NEC assessed by clinical criteria without radiological or pathological confirmation were:
a history consistent with NEC and presence of a palpable abdominal mass associated with overlying abdominal wall cellulitis pneumatosis intestinalis, portal vein gas or presence of a fixed dilated loop of bowel on serial examinations. Radiographic diagnosis made by 2 radiologists unaware of treatment or infant identity suspect NEC classified as cases with clinical history of NEC but no confirmatory radiological, surgical or pathological results
|
| Notes |
Sample size estimation done
10 neonates out of 43 cases of definite NEC in the treatment group and 12 out of 41 cases in control group did not receive any trial solution prior to diagnosis |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
Randomisation using a list of random numbers with 5 IgG and placebo neonates in each block of 10 envelopes |
| Allocation concealment (selection bias) |
Low risk |
The hospital pharmacists provided ampoules of trial solution identified only by name |
| Blinding of participants and personnel (performance bias)
All outcomes |
Low risk |
Blinding of participants and personnel. The trial solutions were packaged in identical glass snap‐top vials |
| Blinding of outcome assessment (detection bias)
All outcomes |
Low risk |
The X‐rays of all infants diagnosed with NEC were reviewed independently by 2 radiologists who were unaware of the treatment group or identity of the infants. X‐rays of infants in which there was diagnostic discordance between the 2 radiologists were jointly reviewed and a consensus reached |
| Incomplete outcome data (attrition bias)
All outcomes |
High risk |
Intolerance to feeds, consent withdrawal and protocol errors delayed study treatment. Thus, 26% of neonates who subsequently developed NEC had not received study medication |
| Selective reporting (reporting bias) |
Unclear risk |
We were unable to obtain the study protocol |
