Gschließer A, et al. Inter-expert and Intra-expert Agreement on the Diagnosis and Treatment of Retinopathy of Prematurity. American Journal of Ophthalmology. 2015;160(3):553-60.e3. The investigators showed 52 retinal photographs to 7 expert ophthalmologists, and then 8 weeks later showed them to the same experts again.

There was little agreement in staging, on the presence or absence of plus disease, on whether there was aggressive posterior disease, on whether the patient needed treatment (using the ETROP) criteria, or on what kind of treatment they would offer. The Kappa agreements were only between about 0.23 and 0.4, between different experts. Even when they looked at the consistency of the scores, by the same experts, they found quite a lot of variability, with Kappa scores mostly between 0.5 and 0.6.

I wasn’t aware of the previous studies that this group refer to in the discussion section of this worrying article. But they describe 3 other studies that show similar things. It seems that there is a lot of variability in the other studies also in terms of diagnosing whether an infant is truly at threshold (needing treatment) there is also disagreement on defining the limits of zone 1. This new study used wider angle retcam images, compared to the previous ones.

Post ED, et al. Milk production after preterm, late preterm and term delivery; effects of different breast pump suction patterns. J Perinatol. 2015. This study compared the efficacy of two different models of breast pump. Just over 60 mothers and babies, (of preterm, late preterm and term infants), who needed to express their milk either pumped with a “sucking” action, and a newer version which generates an irregular suction pattern: it produces periods of 60, 90 or 120 sucks per minute, as well as pauses in suction randomly distributed in time. Mothers were assigned to one group or the other based on equipment availability. There are a couple of things in the report that I don’t understand, they say that mothers were only included if they were pump dependent for at least 7 days, and expressed at least 7 times a day. But they received the equipment when they started expressing, and one of the outcomes was recorded usually before 7 days post-partum. I guess it might mean that there were other mothers who dropped out after starting to pump with one of the 2 devices, but they never got to 7 days, or didn’t pump often enough, so their data were not included. This is a potential bias in the study, but I am not really clear on this.

The mothers who used the irregularly pumping breast pump achieved a 50 mL in a day breast milk production more than 24 hours earlier. More of them were able to produce more than 500 mL a day, and they produced 500 mL or more about 2 days earlier than the comparison group.

Even the controls had a pumping suction pattern which was not regular, they started with a 2 minute stimulation phase of 120 sucks per minute, followed by the expression phase of 60 per minute. Which is already much more like natural sucking than the standard breast pumps we have available, that maintain the same sucking action and frequency throughout, unless you manually vary the frequency. A previous RCT Meier PP, et al. Breast pump suction patterns that mimic the human infant during breastfeeding: greater milk output in less time spent pumping for breast pump-dependent mothers with premature infants. J Perinatol. 2012;32(2):103-10. that I also was not aware of, already showed a benefit of the irregular suction patterns in mothers of preterm babies, the new study with a less rigorous design, suggests the same benefits for milk production in mothers of larger preterm, or term babies.