• plasma levels of vitamin A at birth, 7 and 28 days [ Time Frame: birth, 7 and 28 days ] [ Designated as safety issue: Yes ]
• hepatic stores of vitamin A at 36 corrected weeks [ Time Frame: 36 corrected weeks ] [ Designated as safety issue: No ]

• conjunctival impression cytology (CIC)
• plasma levels of vitamin A at birth, 7 and 28 days
• hepatic stores of vitamin A at 36 corrected weeks

Vitamin A is important for the development of healthy eyes and lungs. Very low birth weight premature babies have low body stores of vitamin A and are prone to diseases of the eye and lungs. Previous work has shown that intramuscular (IM) vitamin A reduces the number of babies who require prolonged oxygen therapy, and may also reduce the number of babies affected by retinopathy of prematurity (ROP)). There is also some evidence that the conjunctiva shows signs of deficiency of vitamin A in premature infants, particularly those who develop ROP. Our own work here in Glasgow suggests that, compared to babies born at full term, premature babies' eyes are less sensitive to light and we believe that this may reflect shortage of vitamin A in the eye. This study will examine the effects upon the eye of giving extra intramuscular vitamin A to very low birth weight, premature infants. We will also measure blood levels of vitamin A and calculate liver stores of this nutrient.

Eligible infant will be those infants born at < 32 completed weeks gestation and/or weighing < 1501 grams birth weight who have been admitted to either Princess Royal Maternity or Queen Mother's Hospital within the first 24 hours of life. If informed, written consent is obtained within 48-72 hours of birth, the infant will be randomised into either control or intervention group.

The intervention group will receive IM vitamin A (Aquasol A)10,000IU three times weekly; control infants will receive mock injections. Injections will be continued for 4 weeks (maximum 12 injections). If enteral feeds are tolerated (defined as more than 75% of predicted intake via the enteral route)after the 14th day, oral vitamin A (as part of a multivitamin preparation) will be commenced and IM vitamin A discontinued. The dose of oral vitamin A will be 5000IU daily (= 0.6ml Dalivit), continued through discharge from the neonatal unit until the first birthday. The same oral vitamin supplement will be given to all VLBW babies, whether or not enrolled in this study. For infants receiving parenteral nutrition, Vitlipid N infant (4ml/kg/day) will be commenced on day 2, or at the discretion of the attending neonatologist. This will be given in addition to IM vitamin A.

The study design is partially blinded whereby control infants will have mock injections (as described by Tyson et al.), rather than placebo injections. Infants randomised to placebo will simply have a sticking plaster applied to a leg prior to the screens being withdrawn. The research nurse will therefore be blinded to the infant's randomisation.

Blood samples will be collected from enrolled infants at birth (or immediately after randomisation), on day 7, day 28 and at 36 corrected weeks. Samples will be separated, frozen and plasma retinol subsequently analysed by high pressure liquid chromatography.

The RDR test will be performed as close as practicable to 36 corrected weeks, and whenever possible in conjunction with routine blood sampling. The baby will be given oral vitamin A, 2000IU/kg, and a second specimen of blood obtained 3 hours after administration of vitamin A. As well as measurement of plasma retinol concentration, red blood cells will be analysed for the DHA content of the cell membrane.

Retinal function will be assessed using the electroretinogram (ERG), in conjunction with routine ROP screening and as close as possible to 36 corrected weeks. The ERG luminance-response function will be recorded using different filters and background lighting to distinguish rod and cone responses. Conjunctival impression cytology (CIC) will be performed coincident with the ERG by taking a single sample from the bulbar conjunctiva, using a Millicell® filter.

All infants will be examined weekly for signs of vitamin A toxicity, including mucocutaneous lesions, bone and joint abnormalities and fullness of the anterior fontanelle. Weekly blood tests during the period of IM injections will include full blood count and liver function.

• Preterm Birth
• Retinopathy of Prematurity

• Drug: Aquasol A
• Drug: aquasol A
• Other: sham injection

Inclusion Criteria:

• Infants born at < 32 completed weeks gestation and/or weighing < 1501 grams birth weight who have been admitted to either Princess Royal Maternity or Queen Mother's Hospital within the first 24 hours of life.

Exclusion Criteria:

• Congenital ocular abnormality