Neonatal Research

Delayed cord clamping or cord milking for the very preterm newborn… or both?

What should we be doing, given the current state of the evidence, for the very preterm neonate?

I think the evidence is now pretty clear that full-term babies have benefits, and no significant harm, from delayed cord clamping, which has been for a defined period of time (1 minute, 2 minutes, or 3 minutes) or until after a defined clinical event, such as when the cord stops pulsating (sometimes with a maximum duration, of for example 5 minutes) or when the baby is breathing well, or after placental descent, in the various studies. The major benefit in the term baby is an improved (more physiologically normal?) iron status and hemoglobin. **this paragraph has been updated, see note after post**

In larger preterm babies the same is probably true. Studies confined to this group have been positive.

In the very preterm baby the evidence as summarized in the latest Cochrane review was still lacking. This group of babies, with of course, the highest risk of complications, must be investigated as a separate group. Potential benefits and possible harms may well be different.

The latest Cochrane review suggests there may be a reduction in all grades of IVH, but not clearly an effect on severe IVH, there may be a reduction in NEC but the confidence intervals are wide, and surgical NEC is not clearly affected, there may be “better circulatory stability” but what this means for outcomes is not certain.

There have been no significant harms shown from either delayed clamping or cord milking, but of course the precise way in which the practice is performed in the studies, and the proportion of randomized infants who actually receive the intervention, and the reasons for, and outcomes of, not following through with the randomized intervention will be different between studies, and have to be understood before we can figure out what to do.

There is a lot going on in this field right now, so its difficult to stay up with everything, at the PAS-meeting this year there were many studies, ancillary studies and physiologic investigations that were relevant. I haven’t had time to digest them all.

One important factor to consider is that the physiologic benefits demonstrated, in animal models, of delaying cord clamping are not due solely (or even mostly) to transfusion effects. Cardiovascular adaptation around birth is different when the cord is clamped after the onset of breathing, I am not sure if there is a similar study of the effects of cord milking, but I would guess that the effects would be quite different. I don’t think we should assume that the two procedures are equivalent, even if the same amount of extra blood is delivered.

Let’s look at some of the studies published since the last Cochrane review.

Elimian A, et al. Immediate compared with delayed cord clamping in the preterm neonate: a randomized controlled trial. Obstetrics and gynecology. 2014;124(6):1075-9. This study randomized 200 mothers of 24 to 34 weeks gestation to either immediate or delayed (30 seconds) clamping. The delayed clamping group were also allowed to get cord milking (2 to 3 times). The primary outcome was the need for transfusion. Which personally I don’t care about, on the other hand I understand the need to sometimes use intermediate outcomes, especially in modestly sized single center trials. Anyway in this trial they found no differences of any importance between the groups, but the average gestational age in the two groups was 31 weeks, so most of the babies would be expected to do well. What I found most surprising in this study was a rate of about 25% of transfusions (at least one transfusion) in babies who were relatively mature. There was also a surprisingly high rate of intraventricular hemorrhage, 20% of the immediate clamping group, and 11% of the delayed (most of course being grade 1 or 2). Overall, no substantial improvements in clinical outcomes, but all in the direction of benefit of delayed clamping for 30 seconds in association with cord milking. For example there was 1 NEC in the delayed group, and 3 in the immediate group.

Krueger MS, et al. Delayed cord clamping with and without cord stripping: a prospective randomized trial of preterm neonates. Am J Obstet Gynecol. 2015;212(3):394 e1-5. Compared 67 babies, 22 weeks to 32 weeks gestation, randomized to either 30 seconds of delayed clamping, or delayed clamping combined with cord milking. They found no added benefit of the milking, specifically no improvement in hemoglobin.

Alan S, et al. Effects of umbilical cord milking on the need for packed red blood cell transfusions and early neonatal hemodynamic adaptation in preterm infants born <!–=1500 g: a prospective, randomized, controlled trial. J Pediatr Hematol Oncol. 2014;36(8):e493-8. Forty-eight VLBW infants of under 32 weeks randomized, to either immediate clamping, or milking, performed 3 times, before the cord was clamped. The primary outcome was the number of the transfusions required, which was not affected, it is hard to make much sense of the other results, due to how they are reported.

March MI, et al. The effects of umbilical cord milking in extremely preterm infants: a randomized controlled trial. J Perinatol. 2013. Seventy preterm infants 24 to 28 weeks gestation were randomized to milking twice before cord clamping, or immediate clamping. Primary outcome was need for transfusion. There were significantly fewer IVHs, and fewer NEC cases (not statistically significant). I can’t understand the severe IVH numbers, as there seem to be more serious IVH than the total numbers of IVH.

Patel S, et al. Effect of Umbilical Cord Milking on Morbidity and Survival in Extremely Low Gestational Age Neonates. Am J Obstet Gynecol. 2014. This is the largest of these studies, with over 300 babies in a before and after study, the cord milking period included the usual 30 second delayed clamping with 3 episodes of milking. Survival, IVH, severe IVH and NEC were all improved in the milking period compared to the previous period.

Katheria AC, et al. The Effects of Umbilical Cord Milking on Hemodynamics and Neonatal Outcomes in Premature Neonates. The Journal of pediatrics. 2014. Anup Katheria has published a couple of papers from this study, this is the one with the clinical outcomes, 60 babies under 32 weeks were randomized, the milking group had the cord milked twice by the obstetricians before clamping, compared to immediate clamping. The primary outcome was the SVC flow, which was around 20% higher in the umbilical milking group. There seemed to be fewer severe IVH and less BPD with cord milking, and no NEC results are given.

Hosono S, et al. One-time umbilical cord milking after cord cutting has same effectiveness as multiple-time umbilical cord milking in infants born at <29 weeks of gestation: a retrospective study. J Perinatol. 2015. In this paper from the group who first reported an RCT of cord milking, they compared a practice of milking the cord after clamping (from 2007 to 2008) to the practice as described before, milking by the obstetrician before clamping, (2001-2002) I don’t know about all the other missing years, and there were only 20 babies in each group. There was similar effects of the 2 procedures in terms of cardiovascular adaptation and hemoglobins.

I think we still need to remember that multiple small trials may inflate the size of any benefit, and the APTS trial, and other large trials, are needed to be sure that the benefits are real, but it looks quite unlikely that delaying cord clamping or cord milking are harmful. I guess the next stage will have to be trials comparing delayed clamping (probably for 1 to 2 minutes) to cord milking (combined with 30 seconds maximum of delayed clamping), to perhaps milking after clamping.


**In response to the comment by Ola Andersson below, I corrected the paragraph about delayed cord clamping in the term baby, and I have also added a link to the Cochrane Review. Previously the paragraph implied that the maximum duration of cord clamping studies was 2 minutes, as I had in my head a couple of large studies, but other large studies have had longer durations, which might possibly be needed to get the full benefit**