Neonatal Research

Infants with severe neurological impairment may have difficulty with oral feeding, and are at risk of aspiration if they regurgitate. Therefore when a feeding gastrostomy is performed, there is often an accompanying fundoplication, It even becomes sometimes so banal that the term ‘gastro-fundo’ may be used (I say banal not to imply that the implications are minimized by the medical and nursing staff, but it has become almost routine for a subgroup of infants).

But is it a good idea? In such infants, if they need a gastrostomy should we routinely perform a fundoplication? Does fundoplication improve outcomes, reduce regurgitation, prevent re-admission, prevent aspiration, improve quality of life?

I used to think that fundoplication was a clearly proven therapy for preventing regurgitation, and therefore it must be a good thing for infants at risk of aspiration. This new study evaluated the records of over 4000 infants with impairments and a gastrostomy who either did or did not have a fundoplication at the same time. It ends with the following paragraph summarizing their results:

Infants with neurological impairment who underwent fundoplication at the time of GT placement did not have a reduced rate of reflux-related hospitalizations during the first year compared with those who underwent GT placement alone, despite propensity score matching. This may be due to a lack of effectiveness of fundoplication in preventing these complications or due to differences in the patient groups that were inadequately accounted for in the matching.

I think the final sentence is absolutely correct, these data show no benefit of fundoplication, but why? We need to know, so that these vulnerable children can get the best interventions targeted to improving their outcomes, and their, and their families’, quality of life. .

The young leaders’ circle is a charitable organisation that is a partner with the Sainte-Justine Foundation in the winter triathlon fund-raiser that is taking place on February the 7th in Montreal.

The funds raised will be going to support maternal and neonatal care at Sainte-Justine, and  specifically a center of excellence in Neonatal Intensive Care. We have a team from the neonatal intensive care unit who will be participating. We have only just registered our team, so we don’t yet have many sponsors, to reach our goal we need contributions in the name of néonatologie Ste-Justine.

If you want to throw in a few dollars (or a lot of dollars) you can do so on-line, so it won’t take more than a couple of minutes. Click on the link and give to a good cause. English version here or in French at http://www.triathlon-sainte-justine.org/page/neonatalogie_ste_justine

I didn’t mention this article when it first came out as it is so obviously nonsense. It came back to my attention as there is a really well done deconstruction on the blog ‘Science Based Medicine‘ which I strongly recommend as a regular web destination. I learnt from this new post that the AAP have had the godawful idea to include the publication in their AAP grand rounds!

This is a profoundly flawed RCT of just over a 100 preterm infants (between 28 and 37 weeks gestation) randomized to either usual care or wishful thinking. Many of the flaws are well described on Science Based Medicine. One flaw which is not discussed there is the failure to do an Intention to Treat analysis, in fact there were 8 babies of 55 in the OMT (Osteopathic Manipulative Therapy) group and 1 in the controls that were not analyzed for various reasons. The babies had an average gestational age of 34 weeks, but nevertheless a mean length of stay of 31 days in the controls, which is far longer than 34 week babies stay in my NICU. Supposedly the length of stay was decreased to 26 by messing about with them, but an easier way to reduce length of stay is to just delete the data from some babies who had long stays!

Most importantly Osteopathic Manipulative Therapy is nonsense. It is pseudo-scientific hocus-pocus, and subjecting fragile preterm babies to it is unethical. Why BMC Pediatrics published this and why on earth the AAP have given it a wider readership is completely unclear to me. Next they will be promoting ear candling and homeopathy, or suggesting that immunization causes autism.

Hallenberger A, Poets CF, Horn W, Seyfang A, Urschitz MS: Closed-loop automatic oxygen control (clac) in preterm infants: A randomized controlled trial. Pediatrics 2014. This is a report of a small multicenter study of the use of a system built into a specific ventilator to control oxygen administration based on pulse oximeter signals. The authors were from 4 NICUs, which all interestingly had different saturation target ranges, the lower limits were 80, 83, 84 and 90% saturation (which is interesting in view of the criticisms of the oxygen targeting trials!) in a crossover design they examined how much time the infants were in there target range when using the closed-loop system compared to standard practice, 24 hours in each mode. The infants spent more time in the desired range with the system than without, and, unlike the other published trial using a different system, and different target ranges, this study showed less time below the target range with the system. The bedside nurses had to do fewer manual adjustments of the FiO2 as a result, which is a valuable goal in itself.

Joy R, Krishnamurthy S, Bethou A, Rajappa M, Ananthanarayanan PH, Bhat BV: Early versus late enteral prophylactic iron supplementation in preterm very low birth weight infants: A randomised controlled trial. Archives of disease in childhood Fetal and neonatal edition 2013. 104 VLBW babies were randomized to receive their iron starting at 2 weeks, or starting at 6 weeks. By 12 weeks of age the Hemoglobin was 1 gram/dl higher, ferritin was much higher, and there were no adverse effects seen, specifically no increase in NEC or sepsis, retinopathy or PVL, although clearly there is very little power for these outcomes. This is a well done trial, addressing an important clinical issue, although too small to be certain about safety. One criticism of this trial is that there is no evidence that the authors performed a systematic review prior to starting the trial, and they do not really address what this trial adds the totality of the evidence for the very low birth weight infant regarding early versus late supplementation. I am agreement with Iain Chalmers, who has written about this extensively (Clarke M, Hopewell S, Chalmers I: Clinical trials should begin and end with systematic reviews of relevant evidence: 12 years and waiting. The Lancet 2010, 376(9734):20-21.)  It should be an absolute requirement prior to starting a trial to find the totality of the evidence that exists, and when reporting to update the systematic review with the new evidence. Maybe there is only one previous relevant trial, and that could be stated, in this particular example, there are at least 3. Adding together all of the safety data could be really worthwhile, and would have made this report more helpful for practice.

Binenbaum G: Algorithms for the prediction of retinopathy of prematurity based on postnatal weight gain. Clinics in Perinatology 2013, 40(2):261-270. The latest issue of Clinics in Perinatology is all about Retinopathy, with some excellent reviews. This specific article is a resumé of the information about the importance of postnatal weight gain, adn about some models that have been constructed to predict RoP and determine which babies should be screened. The overall conclusion is that they look promising but we don’t yet have enough data to not screen babies who have low risk scores. The other conclusion should be that early postnatal nutrition is hyper-important, and making efforts to optimize the quantity and the quality of that nutrition will save eyes! With the advice from many sources to avoid saturations below 90% we will likely see more retinopathy, which will make every other effort that we can make to reduce risks even more important.

If you’ve got a few minutes to spend in the virtual company of some of the greats of EBM, the JAMA network have a microsite with videos of interviews largely animated by Richard Smith (former editor of the BMJ) http://ebm.jamanetwork.com  Fairly light viewing, and you can get an insight into how each of them became involved in promoting medicine based on the scientific evidence.

Early Human Development, like some other journals, sometimes publishes conference proceedings.  In October 2009 a supplement was published with articles reflecting presentations made at a conference in Torino, in July 2010 another was published with articles written by presenters at a neonatal nutrition conference. The latest Early Human Development email, that I receive for new publications on-line, includes correction notices for 2 of those articles. This caught my eye as they are both about microbiome/probiotics in preterms, both co-authored by Maka Mshvildadze and Josef Neu. The correction notices both note that entire paragraphs had been copied from previous publications, and had not been appropriately acknowledged. The corrections suggest that they should have been enclosed in quotation marks, but I think honestly that they should not have been copied verbatim at all. 

In addition, with a bit of digging there are clearly other parts of the manuscripts which are also copied. Take this example, from the publication in EHD, (which is not listed in the correction statement as needing quotation marks)

Alterations in diet affect the composition and, more importantly, the collective metabolic output of the microbiota. Thus, deliberate dietary supplementation with bacterial fermentative substrates, usually complex carbohydrates can increase the growth of potentially beneficial microorganisms as well as the production of SCFAs, including butyrate, and indirectly modify immune function [47]. This approach is called “synbiotics.” Obviously, the emerging ability to couple metagenomics of bacterial populations with resultant metabolomics will facilitate rational attempts to manipulate endogenous gut microbiota by dietary changes

Compare this to an earlier article by Andrew Neish

Alterations in diet affect the composition and, more importantly, the collective metabolic output of the microbiota. Thus, deliberate dietary supplementation with bacterial fermentative substrates, usually complex carbohydrates (eg, inulin, oligofructose), can increase luminal concentrations of SCFAs, including butyrate, and indirectly modify immune function….  Prebiotic substrates could be administered with live bacteria most able to exploit that energy source, an approach called “synbiotics.” Obviously, the emerging ability to couple metagenomics of bacterial populations with resultant metabolomics will facilitate rational attempts to manipulate endogenous gut microbiota by dietary changes.

There is a reference to Neish’s article given, but no indication that large parts were directly copied. There are other large portions that are taken from Neu’s own previous work, which has sometimes been called “auto-plagiarism”. I am much more sanguine about that, it is hard to always find a new way to say the same thing, on the other hand the co-authors on the previous publications who actually produced the initial text have a right to be pissed that their work was not appropriately acknowledged. The text of the EHD article directly copies the results section of the abstract of another trial (Rinne M et al ) which is mentioned by name as Rinne et al, but doesn’t actually appear in the references, the reference number is to a different article altogether (that looks like its just a mistake, but the results are quoted verbatim):

Numbers of different types of stools, vomits and crying time were comparable between the groups during the 7th and the 12th weeks of life. Dominant microbiota consisted of bifidobacteria throughout the study. At 6 months, there were less clostridia in feces in the placebo compared with the probiotic group (P = 0.026), whereas after long-term follow-up at 2 years, there were less lactobacilli/enterococci and clostridia in feces in the probiotic group than in the placebo group (P = 0.011 and P = 0.032, respectively), reflecting the impact of clostridia as a marker of microbiota succession in healthy infants.  

Perhaps its not a big deal to directly quote factual information like that, but I have never done it in anything I have written, and the final phrase is clearly an opinion and should not be directly copied without attribution.

In the second of the two corrections a long paragraph lifted from an article in the PNEJM by James Madara is stated to need quotation marks. This is the EHD paragraph

“Components of bacteria provide chemical signals that are recognized by specific receptors – called toll-like receptors (TLRs) – of the innate immune system. The mechanisms of how microbes or their components interact with this system to provide innate immunity are still in the early phases of research. It is assumed that the healthy intestinal surface somehow defuses the threat of commensal bacteria to the lumen where they thus reside undetected. A study by Rakoff-nahoum et al provides insights on how this happens: commensal bacteria interact with the intestinal surface and to some degree trigger TLR-signaling. The authors used mice deficient in either TLR-2 or TLR-4 or a necessary downstream component of the TLR pathway (MyD88), to prevent TLR-signaling. Such mice were shown to have a profoundly exaggerated response to intestinal injury. This effect is not a consequence of acute inflammation or the unrestrained overgrowth of commensal bacteria; rather, it results from the loss of TLR-dependent conditioning that allows the intestinal surface to maintain its normally resistant homeostasis. Surprisingly this interaction is required to maintain the architectural integrity of the intestinal surface. Thus it seems that the epithelial and resident immune cells do not simply tolerate commensal bacteria but are dependent on them.” 

Bu this isn’t exactly a quotation, as the order of some parts have been re-arranged, and other sentences slightly edited, the original is:

 Like all bacteria, they release chemical signals with conserved patterns recognized by specific receptors ― called toll-like receptors (TLRs) ― of the innate immune system. It is therefore assumed that the healthy intestinal surface somehow defuses the threat of commensal bacteria to the lumen, where they thus reside undetected. A recent study by Rakoff-Nahoum and colleagues provides insight into how this happens: commensal bacteria interact with the intestinal surface and, to some degree, trigger TLR signaling…..

Surprisingly, this interaction is actually required to maintain the architectural integrity of the intestinal surface. Thus, it seems that the epithelium and resident immune cells do not simply tolerate commensal bacteria but are dependent on them….

The authors used mice deficient in a necessary downstream component of the TLR pathway, thereby preventing all TLR signaling. Such mice have a profoundly exaggerated response to intestinal injury. This effect is not a consequence of acute inflammation or the unrestrained overgrowth of commensal bacteria; rather, it results from the loss of TLR-dependent conditioning that allows the intestinal surface to maintain its normally resistant homeostasis.

I’m sorry but this is not unintentionally leaving off quotation marks, but this really looks like extensive plagiarism, and I think that just adding quotation marks is not sufficient, these articles should be retracted.