How to measure growth? How to describe growth rates? What does it matter?
Several related articles in today’s post, the first two are about how to measure growth in preterm infants:
It is clear that body weight is not good enough. Although it can be measured easily, reproducibly and precisely, just because body weight is increasing along the wanted percentiles does not mean that growth is optimal: excess fat, and not enough of everything else, is common in preterm infants. We showed in our study that enhanced nutrition can almost completely prevent post-natal growth failure in the preterm infant, when calculated as loss of body weight Z-scores, but that length z-scores still fell between admission and discharge (by a mean of 1.5, compared to 1.7 with the older cohort) . Just looking at our most immature, longest stay, babies you can see that many of them are ‘short and plump’. Head circumference z-scores were maintained, though, and, as a very rough proxy for brain growth, that is re-assuring that the enhanced nutrition allowed good cerebral growth. Also, measuring length is rather inaccurate in usual daily practice, knemometry seems to be more accurate, but I can’t find a source of a device to do it. When I started working at the Royal Victoria Hospital in Montreal they were using a stadiometer, which is more accurate, but requires a lot more disturbance of the babies, so was only done after they were quite stable, and would be difficult to introduce into routine practice elsewhere.
So what else could we do? Body composition measurements would be ideal, but all current methods require either expensive equipment or extensive manipulation of the baby, or both. It would be great if there was some other measure that was shown to correlate with fat-free mass, that you could simply add to weekly weight and head circumference measurement, and that could then be used to evaluate changes in nutritional practice.
In this new publication the use of additional measures, mid-upper arm circumference, and mid-thigh circumference were compared with repeated measures of weight, length, and head circumference. They showed that the measures are simple and reproducible, and seem to have different progression to measures of weight and head circumference. They haven’t yet proven that they are a useful addition to our other measures, but I am hopeful that something this clinical and simple could give new insight into the quality of growth of our babies, rather than just the quantity.
Tanis Fenton is the source of the growth charts that we use in our NICU, she has performed several systematic reviews that have led to our adoption of her charts, and now presents a systematic review of measures of growth velocity calculation. You can present such data as grams per kilogram per day, or as gains in z-scores, or in many different ways, the question of this review was whether or not there is some sort of consensus. After reviewing hundreds of articles they finally state:
The lack of standardization of methods used to calculate preterm infant growth velocity makes comparisons between studies difficult and presents an obstacle to using research results to guide clinical practice.
They also note:
It is important for researchers to identify which growth charts were used to calculate z scores.
Which would seem to be self-evident, but is apparently not. In our article we did refer to (and use) the growth and percentile charts and on-line z-score calculator of Tanis Fenton. All referees should ensure that the source of the growth data are clearly referenced.
Does it matter if we can get preterm babies to grow faster, and stay closer to their percentile? Although this is a small study with only 42 babies included, they were able to show that when babies increase their weight better, they have better cerebellar volumes, even if they are fatter than they would otherwise be, and have higher fat-mass, they still have better cerebellar volumes.
In this study the authors compared brain growth between preterm infants and fetuses who remained in utero at similar gestational ages. They selected infants who did not have obvious brain injury and noted that being ex-utero, rather than in-utero had significant effects on growth of the brain in multiple regions. They corrected for several factors, but I can’t see any adjustment for nutritional intakes, which could potentially account for many of the changes that they have seen.
Finally, an abstract from this year’s PAS meeting with Julie Schneider as first author, from the incomparable Steven Miller and the team at the Sick Kids hospital in Toronto; they also looked at cerebral volumes in various areas of the preterm infant’s brain, as well as fractional inosotropy (which I could probably explain to you, but then I would have to kill you). They saw that the nutritional intakes in the first two weeks of life were strongly associated with improved brain growth and more normal fractional inosotropy when the babies then reached term.
When I was learning neonatology we were just starting to take nutrition seriously, fortunately I had a mentor (Neil Finer) who was already taking nutrition seriously, and the groups I have been involved with since then have been interested in, and committed to, providing optimal nutrition to small preterm infants.
We must continue to improve our standards, find ways to ensure growth that reflects in utero standards, and ensure that growth quality, and not just quantity is optimised to improve short and long-term outcomes.