Blood transfusion and necrotising enterocolitis
| Citation | Study group | Study type (level of evidence) | Outcome | Key results | Comments |
|---|---|---|---|---|---|
| Bednarek et al (1998) | Prospective analysis of blood transfusions and outcomes (including NEC) in 825 very low birth weight (<1500 g) infants in 6 neonatal units over 1 year, with adjustment for birth weight and illness severity. The 6 units were categorized into low, medium and high transfusion units based on the mean number of transfusions per infant | Prospective ecological study (level 2c) | Incidence of NEC | Adjusted OR (95% CI) for the: High transfusing units: 1.1 (0.5–2.2) Medium units: 1 (reference) Low transfusing units: 0.3 (0.1–0.08) p<0.05 The low transfusing NICU group showed a significantly lower incidence of NEC compared with the middle and high transfusing units. | Association difficult to interpret in ecological studies as the association found between transfusion and NEC is at the level of the units but was not studied at the individual neonate level Findings may be related to other practices in the specific NICU (e.g. restricted transfusion policy) or reflect confounding by indication for transfusion (e.g. infants who have NEC may require more transfusions). Time interval between transfusion and NEC not available (any transfusion at any time before NEC was counted) |
| McGrady et al (1987) | Case-control study of 33 neonates with NEC during an outbreak, and 40 controls matched on birth weight, duration of stay in the unit and approximate date of admission. Median birth weight of cases = 1360 grams, median gestational age = 32.5 weeks | Individual case-control study (level 3b) | Risk factors for NEC | Transfusion was highly and significantly associated with NEC, crude OR = 15.5 (95% CI = 2.59–92.51); RR = 8.98 (95% CI = 1.08–74.6) after adjustment for therapy with caffeine, theophylline and furosemide. There was no association with type or timing of feeding | This study was that of an outbreak of NEC and not the endemic form of NEC. Epidemic NEC may be importantly very different from endemic NEC |