I thought I knew, sort of, what to do about iron supplementation in the preterm. Preterm babies outgrew their iron supply, there isn’t very much in breast milk, and you need to supplement to minimize the appearance of anemia of prematurity.

There was previously a trial which compared transfusions and iron status among 204 preterm infants (<1300g birth weight) who were randomized to start iron as soon as they were tolerating 100 mL/kg/d of milk, or to wait until they reached 61 days. The babies received 2 mg/kg/d at first and it was increased to 4 if the hematocrit was below 30. That trial showed fewer transfusions and less iron deficiency with early supplementation. They also published long term follow up, although not powered for that outcome, the 164 babies who were studied (most, 85%, of the survivors) tended to have better outcome in the early group.

There is also a Cochrane review showing improved hematocrits with routine enteral iron prophylaxis in the preterm.

Now a new trial has been published which compared giving 2 mg/kg/day of iron, started when the babies were tolerating 120 mL/kg/d of milk, to a multivitamin preparation without iron. The babies were 150 VLBW (<1500g) infants and the primary outcome was hematocrit at 36 weeks, transfusions were also noted. They note that the control group were receiving around 2 mg/kg/d already in the breast milk fortifier that they were adding to the breast milk. There were no differences in the results between the groups. The authors note that they have a ‘liberal’ transfusion policy, which they say is based on the previous evidence regarding ‘long-term benefits of maintaining higher hemoglobin levels in extremely low birth weight (<1000 g birth weight) infants’. What they don’t tell us is how many transfusions the infants received prior to entering the study. That info would have been helpful: a blood transfusion or two would have given substantial amounts of iron to the babies, and could be a reason for not finding a difference between the groups. that would make it less easy to extrapolate the findings to other NICUs, such as ours, where we have not liberalized our transfusion guidelines.

And just how good is that data about long term benefits of liberal transfusion? Although the authors reference the PINT trial and the Cochrane review, neither of those sources show an improved long term outcome! The PINT study showed slightly better 18 month Bayley scores, differences consistent with chance, p value about 0.09, and only when they looked at the proportion of babies less than 85 (not 70 which was pre-specified as the outcome of interest) did they show a difference that looked more convincing. The other trial that has reported some long term outcomes was the trial by Ed Bell and others, who followed only 56 of their 100 patients to early school age, and some measures of cognitive function were better in the liberal transfusion compared to the conservative group.

In summary I do think it seems possible that some outcomes are better with a higher transfusion threshold, but it certainly isn’t definite, and before using a lot more blood transfusions we should get better data. There are other trials in the works.

In the meantime, with a more restrictive transfusion policy, I think continuing to give iron supplementation early, in addition to what we have in the breast milk fortifier still makes sense.