ECMO: not dead yet

Posted on 20 June 2013 by Keith Barrington

There is a sort of a quadruple entendre in the title of this post: ECMO still exists; ECMO is needed if you are ‘not yet dead’ and have a chance of surviving; ECMO decreases mortality if you need it; but, unfortunately, using ECMO may sometimes just delay death. I haven’t written about ECMO before on this blog, even though I have been very involved at certain times in my past, the frequency of needing ECMO for neonatal respiratory disease has become vanishingly rare in my practice (I haven’t referred a meconium aspiration for ECMO in 5 years). ECMO for diaphragmatic hernias and for postoperative cardiogenic shock, however, is still sometimes life-saving. I have put together a number of very recent publications.

Tabbutt S, Ghanayem N, Ravishankar C, Sleeper LA, Cooper DS, Frank DU, et al. Risk factors for hospital morbidity and mortality after the Norwood procedure: A report from the Pediatric Heart Network Single Ventricle Reconstruction trial. J Thorac Cardiovasc Surg. 2012;144(4):882-95. Not surprisingly, if you need ECMO after a Norwood, then you are more likely to die. Of those in this large multicenter cohort of infants with hypoplastic left heart syndrome (n=nearly 600) who had post-op ECMO the transplant free survival at 12 months was only 26% compared to 75% if you didn’t need ECMO (and didn’t get CPR either). Still 26% compared to zer0… I guess it depends on your viewpoint, and your resources.

Another study in a similar vein (or should I say with a similar heart) is an analysis of ELSO registry data of over 700 babies who had a Norwood and then needed ECMO. The survival overall (to hospital discharge) was 31%. Sherwin ED, Gauvreau K, Scheurer MA, Rycus PT, Salvin JW, Almodovar MC, et al. Extracorporeal membrane oxygenation after stage 1 palliation for hypoplastic left heart syndrome. The Journal of Thoracic and Cardiovascular Surgery. 2012;144(6):1337-43. The authors note that complications of ECMO are a bad risk factor. The median duration of ECMO among survivors was very close to 72 hours. They show that there was an increased duration of ECMO among non-survivors, and that the Odds Ratio for mortality for every additional day of ECMO is 1.2. Useful information for counselling parents. They also note that there were 28 babies who went back on ECMO for a second time, after the first decannulation, only 1 survived. Probably if you re-satisfy criteria to go back on ECMO, unless there is a surgically correctable reason, you should not offer a second course of ECMO.

But what if you have congenital heart disease and a diaphragmatic hernia? Is ECMO a reasonable option? Dyamenahalli U, Morris M, Rycus P, Bhutta AT, Tweddell JS, Prodhan P. Short-Term Outcome of Neonates With Congenital Heart Disease and Diaphragmatic Hernia Treated With Extracorporeal Membrane Oxygenation. The Annals of Thoracic Surgery. 2013;95(4):1373-6 This report from the ELSO registry has surprisingly good outcomes, and some surprises in the results. For infants with a diaphragmatic hernia who needed ECMO, survival to hospital discharge for those who also had hypoplastic left heart syndrome and single-ventricle physiology was 55% (33 of 60), 48% (43 of 89) for ventricular septal defect, and 40% (24 of 60) for coarctation of the aorta patients. Which is  a little weird, infants who had the most severe heart disease had better survival than those with less severe heart disease. I can think of a couple of possible reasons for this. But if you need ECMO for your diaphragmatic hernia (CDH) a VSD is probably not very relevant; it just means that you have a really bad CDH. Whereas if you have a hypoplastic left heart, it is much easier to satisfy ECMO criteria, so you might have less pulmonary hypoplasia, and a better prognosis for short term survival. How you might do after the second stage of the hypoplastic left heart repair is not clear.

Madderom MJ, Toussaint L, van der Cammen-van Zijp MHM, Gischler SJ, Wijnen RMH, Tibboel D, et al. Congenital diaphragmatic hernia with(out) ECMO: impaired development at 8 years. Archives of Disease in Childhood – Fetal and Neonatal Edition. 2012. This study was a very complete evaluation of 8 year old survivors of CDH repair. The children who needed ECMO in the neonatal period had lower IQ scores than those who survived without. Dutch children are obviously very intelligent, the CDH children who did not need ECMO had a mean IQ of 112. It must be all that cheese or more likely the raw herring ‘Hollandse nieuwe haring’, yum yum. Or maybe living below sea level with their higher PaO2 helps brain development, it certainly makes them tall. The CDH children who had needed ECMO didn’t do quite as well (mean IQ 91) and both groups had troubles with concentration, but overall their quality of life was really good.

If we look at the ELSO registry data for acute neurologic complications (Polito A, Barrett CS, Wypij D, Rycus PT, Netto R, Cogo PE, et al. Neurologic complications in neonates supported with extracorporeal membrane oxygenation. An analysis of ELSO registry data. Intensive care medicine. 2013:1-8) we find that they are common, occurring in about 20% of supported infants, and associated with increased mortality. As always in database information, we don’t know if that might be because the medical care changed as a result of the complication. Is the increased mortality because care was redirected after a complication?…

Madderom MJ, Reuser JJ, Utens EM, van Rosmalen J, Raets M, Govaert P, et al. Neurodevelopmental, educational and behavioral outcome at 8 years after neonatal ECMO: a nationwide multicenter study. Intensive care medicine. 2013:1-10.     135 Dutch babies required ECMO in a 5 year period. Eight years later the survivors are doing really well.   The mean IQ is 100 ( I know for a Dutch child that is probably rather low, but it still sounds quite good for the rest of the world). They did have some problems, with some attention difficulties, and a few more having learning disability that will probably require specific educational attention. But these outcomes are very encouraging, and in addition, they aren’t dead yet!

Posted in Neonatal Research | Tagged , , | 7 Comments

Speech is developing even before the speech centers appear

Posted on 18 June 2013 by Keith Barrington

Someone who is more expert in this field than me could probably take apart that title, but the message of this article is that, even at 29 weeks gestation, , the preterm brain reacts differently to different phonemes (ba versus ga) and different voices (masculine versus feminine). That’s cool. Because at 29 weeks the neurons which will eventually form the centers involved in speech are still not at their final locations. The brain is still very immature at that point, and those parts of the brain that we usually expect to respond to speech have not been properly constituted. Language really is hard-wired into our brains.

Mahmoudzadeh M, Dehaene-Lambertz G, Fournier M, Kongolo G, Goudjil S, Dubois J, et al. Syllabic discrimination in premature human infants prior to complete formation of cortical layers. Proceedings of the National Academy of Sciences. 2013;110(12):4846-51.

If you take those preterm babies when they are adults, they seem to be processing language with different parts of the brain. Although their scores on standardized testing of language skills were the same as controls born at term, the preterm born adults were using pathways in their cerebellum to process language, but not the term born adults. Constable RT, Vohr BR, Scheinost D, Benjamin JR, Fulbright RK, Lacadie C, et al. A left cerebellar pathway mediates language in prematurely-born young adults. NeuroImage. 2013;64(0):371-8. This seems to me to be new evidence of the plasticity of the brain, which develops differently if you are born very preterm, the average gestation of the subjects was 28 weeks.

Posted in Neonatal Research | Tagged | 4 Comments

We can stop routine suctioning at birth!

Posted on 17 June 2013 by Keith Barrington

I never understood why it was thought to be so important, (indeed by some essential) to stick some sort of suctioning device into the babies mouth, usually right back to the posterior pharyngeal wall, and suck out amniotic fluid. It was as if we thought that despite many millions of years of evolution, human beings were not capable of surviving unless someone sucked out their upper airways after birth. Most mammals do not have access to suction devices at birth, yet surprisingly most of them survive regardless. Fortunately Wally Carlo’s team (does he ever sleep?) have just shown that babies who only have their face wiped (which I guess is analogous to the facial licking that a lot of mammals do) have a clinical outcome that is just as good as babies who get induced vagal stimulation oropharyngeal suctioning by caregivers. We already know that routine suctioning increases bradycardia, now we know that it doesn’t have any routine advantages. Let’s stop. Kelleher J, Bhat R, Salas AA, Addis D, Mills EC, Mallick H, et al. Oronasopharyngeal suction versus wiping of the mouth and nose at birth: a randomised equivalency trial. The Lancet. 2013(0). Of note the current recommendations from several bodies have already abandoned routine suctioning in infants born at term with clear amniotic fluid, but that recommendation was based on very little data, this new trial gives a lot more force to that recommendation. I think we can move from not recommending routine suctioning, to recommending against it.

Posted in Neonatal Research | Tagged , | 3 Comments

Neonatal Updates #35

Posted on 15 June 2013 by Keith Barrington

Huang SS, Septimus E, Kleinman K, Moody J, Hickok J, Avery TR, et al. Targeted versus Universal Decolonization to Prevent ICU Infection. New England Journal of Medicine. 2013;368(24):2255-65. A cluster randomized study in adults in intensive care. There were 3 groups. 1. Screening for MRSA and isolation of positive patients (which was the standard of care before the study). 2. Screening for MRSA and treatment of positive adults with intra-nasal mupirocin and cleaning them with chlorhexidine wipes, for a 5-day period. 3. No screening, just treating all admitted adults with mupirocin for 5 days and chlorhexidine wipes for the entire ICU stay. The primary outcome was MRSA positive cultures (which I think includes for example positive nasal cultures after the first screening culture), and secondary outcomes included all serious nosocomial infections, MRSA bacteremia and bacteremia from all organisms. Universal treatment and body washes reduced serious nosocomial infections much more than the other approaches, both MRSA septicemia and all cause septicemia, by about 44%. It is starting to look like these chlorhexidine impregnated washcloths really work to reduce serious nosocomial infections. We need more studies to see  in which babies it would be safe to do this (by examining absorption and toxicity in preterm and very preterm infants), and we need to restrict their use to the ICUs, to avoid the evolution of resistance, which will surely, eventually, happen if they are too widely used (in domestic animals for example!)

Bertini G, Elia S, Ceciarini F, Dani C. Reduction of catheter-related bloodstream infections in preterm infants by the use of catheters with the AgION antimicrobial system. Early Human Development. 2013;89(1):21-5. Another RCT for infection control, this time in newborn infants, 86 of whom were randomized to get an umbilical venous catheter that was impregnated with an antiseptic silver compound, or control. There was a substantial, and significant reduction in infection related to the catheter with the silver impregnated catheter, but the control rate was extraordinarily high, 22% of the control group had a catheter related infection, during a mean catheter duration of about 8 days. This will need to be tested in NICUs with much lower UVC infection rates, and we really need to be sure that the exposure to silver zeolite is safe.

Boss RD, Kinsman HI, Donohue PK. Health-related quality of life for infants in the neonatal intensive care unit. J Perinatol. 2012;32(12):901-6. This interesting piece is a review of publications of quality of life, the main thrust of the article is that we have no data about quality of life of babies while they are still in the NICU.  They refer to the article that Antoine Payot and I wrote about quality of life of former preterm infants (and infants with other chronic conditions) throughout life, and make a plea to establish quality of life scales for babies while they are still in the NICU. I found that an interesting idea that I had not really considered, we have ways of measuring pain, and several other scores which are peripherally related to QoL, but no agreed method for analyzing QoL during our NICU treatment, which may last months, and, unfortunately, for some babies is the only life they ever know.

Posted in Neonatal Research | Leave a comment

Hyperinsulinemic babies don’t neccesarily have high insulin??

Posted on 14 June 2013 by Keith Barrington

It has been known for some time that some SGA babies with hypoglycemia are hyperinsulinemic. As many as 20% of some small series have shown this phenomenon, they can be suspected because of the high glucose requirements, and confirmed, I thought, but measuring insulin concentrations during a hypoglycemia. I now find out that sometimes babies who have hyperinsulinemic hypoglycemia aren’t all hyperinsulinemic when they are hypoglycemic. (try saying that quickly after a glass of wine).

I have seen this in practice, a baby who needs a lot of sugar, is probably hyperinsulinemic but has low or undetectable insulin when hypoglycemic, I didn’t know previously how to interpret it, but it seems that it is not uncommon. I guess the insulin level is too high when the sugar is normal, but when they get hypoglycemic they are able to suppress insulin production to below the sensitivity of the lab tests.

In these circumstances other indications of hyperinsulinism, such as very low free fatty acids, low beta-hydroxybutyrate, or an exaggerated glucose response to glucagon can confirm the diagnosis. (Hoe FM, Thornton PS, Wanner LA, Steinkrauss L, Simmons RA, Stanley CA: Clinical features and insulin regulation in infants with a syndrome of prolonged neonatal hyperinsulinism. The Journal of pediatrics 2006, 148(2):207-212.)

This new publication (Arya VB, Flanagan SE, Kumaran A, Shield JP, Ellard S, Hussain K, Kapoor RR: Clinical and molecular characterisation of hyperinsulinaemic hypoglycaemia in infants born small-for-gestational age. Archives of Disease in Childhood – Fetal and Neonatal Edition 2013.) examined whether there were abnormalities in the genes regulating insulin secretion, genes that may be abnormal in the permanent types of hyperinsulinemic hypoglycemia. There weren’t.

In these SGA babies with the condition it is transient, although it may last some months, and the new publication lets you know that a) these babies respond well to diazoxide, and b) when you can get the diazoxide dose below 5 mg/kg/day you can safely stop it.

So we need to measure these other indicators of hyperinsulinism if a hypoglycemic baby who needs a high glucose infusion rate needs further diagnostic testing.

Posted in Neonatal Research | Tagged | Leave a comment

Xrays may not be innocuous

Posted on 13 June 2013 by Keith Barrington

This worrying article from Australia of a matched database study shows a significant increase in cancer risk with a CT scan performed during childhood or adolescence, (Mathews JD, Forsythe AV, Brady Z, Butler MW, Goergen SK, Byrnes GB, Giles GG, Wallace AB, Anderson PR, Guiver TA et al: Cancer risk in 680 000 people exposed to computed tomography scans in childhood or adolescence: data linkage study of 11 million Australians. BMJ 2013, 346(may21 1):f2360-f2360.) The increase in incidence is about 24% over the population baseline risk, which is not huge for a relatively infrequent group of diseases, but the follow up of the population was just under 10 years on average, so it might increase as time goes on. That’s about 10 cancers per 100,000 patient years at risk, so don’t panic yet!

Premature babies have a lot of radiographs, and get exposed to a lot of radiation. Donadieu J, Zeghnoun A, Roudier C, Maccia C, Pirard P, Andre C, Adamsbaum C, Kalifa G, Legmann P, Jarreau P-H: Cumulative effective doses delivered by radiographs to preterm infants in a neonatal intensive care unit. Pediatrics 2006, 117(3):882-888. But most of them are way below what you get from a CT scan, even if you add up all the radiation during their entire stay. A few get to quite high levels however, and some of them do end up with a CT scan. Need to consider carefully each time before a CT scan if the very small increase in risk of cancer is worth the clinical information that we will get.

Posted in Neonatal Research | Leave a comment

When she died there was a hailstorm, it was as if the Angels cried.

Posted on 11 June 2013 by Keith Barrington

Evy Kristine (the little girl that Siri Berg, her mother, wrote about) has a website, with a page in English. The title of this post is taken from one of the two very moving poems you can find there written by her elder brother. You can also find a link to a youtube video that the family made, and to other resources.

One of those resources recounts a much more positive interaction with the medical system, showing that we can sometimes get it right. Daniel’s book recounts a prenatal diagnosis of trisomy 13, and multiple interactions with helpful compassionate people who were willing to listen to the family’s wishes, to offer a range of possible approaches, and to support them in the decisions that they made. Daniel lived just 32 hours.

It sounds as if many of them had read John Carey’s article from last year, or maybe (as that isn’t possible) they were just caring people. Dr Carey’s article ‘‘Carey JC: Perspectives on the care and management of infants with trisomy 18 and trisomy 13: Striving for balance. Current Opinion in Pediatrics 2012, 24(6):672-678” is a review of the literature, and ends with recommendations, which I quote in a shortened form from the abstract:

The author recommends a balanced approach to counseling families of the newborn with trisomy 18 and 13 at the time of diagnosis. The counseling process should include presentation of accurate survival figures, avoidance of language that assumes outcome, communication of developmental outcome that does not presuppose perception of quality of life, and respect for the family’s choice, whether it be comfort care or intervention.

Posted in Advocating for impaired children | Tagged | Leave a comment

Neonatal Updates #34

Posted on 8 June 2013 by Keith Barrington

Hartling L, et al. Benefits and Harms of Treating Gestational Diabetes Mellitus: A Systematic Review and Meta-analysis for the U.S. Preventive Services Task Force. Ann Intern Med. 2013 (open access). A very comprehensive review, as is usual from this source, I can’t pretend to have read it all in detail, but the conclusions are: Treating GDM reduces macrosomia (which in turn leads to less shoulder dystocia), but not neonatal hypoglycemia, there is no real evidence of harm.

Ishii N, Kono Y, Yonemoto N, Kusuda S, Fujimura M, for the Neonatal Research Network J. Outcomes of Infants Born at 22 and 23 Weeks’ Gestation. Pediatrics. 2013. 36% of the infants born alive at 22 to 22 weeks 6 days survived to 3 years of age, and 63% of those born at 23 weeks. The rates of severe CP and moderate to severe developmental delay at 3 years were high among the 22 and 23 week infants. The survival rates are very impressive, we need to find ways to reduce impairments.  They also present outcomes for more mature infants up to 25 weeks, and some summary data about in-hospital complications. The rates of BPD are incredibly low, only 30% at 23 weeks, severe intracranial hemorrhage was common at 22 and 23 weeks, over 20%, and a very high proportion had RoP requiring therapy, 30% right up to 25 weeks. This suggests some ways to reduce impairments, I don’t know if they use prophylactic indomethacin, but that intervention reduces severe hemorrhage, and PDA ligation without adverse effects, in a population with such a high risk of severe bleeds, that would be worth considering. Secondary analysis of the TIPP data showed in one high risk group (boys) that developmental outcomes were better with indomethacin treatment. Overall, it is true there was no effect on developmental delay, but the incidence of severe bleeds in the controls was much lower than this, 13%.

Corvaglia L, Aceti A, Mariani E, Legnani E, Ferlini M, Raffaeli G, et al. Lack of efficacy of a starch-thickened preterm formula on gastro-oesophageal reflux in preterm infants: a pilot study. The journal of maternal-fetal & neonatal medicine. 2012;25(12):2735-8. A cross-over study of a new formula with amylopectin. There were fewer reflux episodes, but no difference in total acid exposure of the lower esophagus, and no effect on non-acid reflux. One could ask ‘why bother?’  They start the article stating 

Gastro-oesophageal reflux (GOR) is commonly diagnosed in preterm infants in neonatal intensive care units (NICUs); it is known to cause morbidity, prolong hospital stay and is associated with some relevant complications, such as aspiration of gastric content and oesophagitis.

I beg to differ. GOR is commonly diagnosed, yes: ‘known to cause morbidity’, little or no evidence. The authors state that they enrolled infants with ‘symptomatic’ reflux; the list of symptoms they give  ‘frequent regurgitations, post-prandial desaturations, chewing, hiccupping and back arching’ are not statistically more frequent in infants with more frequent GOR on esophageal impedance/pH monitoring than in infants with less frequent reflux.

Posted in Neonatal Research | 2 Comments

‘Why were they in such a hurry to see her die?’

Posted on 7 June 2013 by Keith Barrington

That quotation is the title of a touching and thought provoking article just published (Berg SF, Paulsen OG, Carter BS. Why Were They in Such a Hurry to See Her Die? American Journal of Hospice and Palliative Medicine. 2013;30(4):406-8.) It is the personal story of a couple who had a baby with trisomy 18. Their individual history mirrors much of what Annie Janvier, Barb Farlow and Ben Wilfond described in their survey of many hundreds of families. It relates the lack of comprehension of many that they did not want an abortion, and the sometimes outright refusal to consider any options in the care of their baby.

At one point the couple go to see a neonatologist to discuss options of care: this was the response ‘The physician gave us the impression that this monster should have been removed long ago and that she was as good as dead.’

The most disturbing part of the story occurs when Evy is born. Despite an agreement to provide modest resuscitation if she doesn’t breathe immediately, no-one intervenes and the father, an intensive care physician/anesthesiologist, has to take her and do the resuscitation himself.

The parents describe how much the 3 days of Evy’s life affected them for the good, which helps us to remember that the value of a life cannot be measured in hours.

The story is accompanied by a discussion by Brian Carter, in which he addresses the issues, and the need to individualize our counseling and decision -making to the desires and values of the family. I think I can be proud that he then uses a list that I created for a previous blog post. It is a list of tips and suggestions to use when talking to parents who have a fetal, or neonatal, diagnosis of a potentially life-limiting condition. It was based on comments from families in Annie Janvier’s paper referred to above, of things that they appreciated, and things that they disliked, in their interaction with physicians, and other care-givers. Unfortunately he improved it a little (and also toned down the language a bit for a respectable journal rather than a blog). So here is the Carter revision of the Barrington tips for being a humane doctor (or other caregiver) in that situation.

(You can click to enlarge)

Barrington checklist
Posted in Advocating for impaired children | Tagged , | 7 Comments

Well SUPPORTed

Posted on 6 June 2013 by Keith Barrington

I think the NIH just stole one of my titles, which was so innovative and unexpected, that I don’t believe they just had the same idea, they must read this blog. A new perspective article in the PNEJM, from, among others, Francis Collins (the NIH director), is titled ‘In support of SUPPORT’. I am going to call my lawyers to complain about infringement of my intellectual property rights. As for the content, well it is strong statement in support of er.. SUPPORT.

A large group of people, ethicists, pediatricians and neonatologists also signed a letter just published in PNEJM, that was called just ‘OHRP and support’, and, as I was one of the signatories, I can’t really complain about that title. But the letter could be considered auto-plagiarism, as it repeats something that I already wrote elsewhere, i.e. ‘Public citizen are full of shit’- no, not really, some of the authors were reluctant, can’t understand why…

It seems that all the public criticism of the OHRP ruling has made them reconsider their actions. OHRP have now issued  a new letter withdrawing the sanctions against the SUPPORT group, and specifically against the IRB in Birmingham. The letter still states that OHRP were right, but they now think that maybe the details of their response were incorrect. Much easier to write that, than an honest letter that said ‘we messed up, and listened too much to poorly informed detractors’.

One thing in that retraction letter that I really disagree with is this idea:

`Doctors are required even in the face of uncertainty to do what they view as being best for their individual patients. Researchers do not have the same obligation’

I was nearly sick. YES WE DO! And it is an obligation that we take very seriously.

I think this comment is based on a very old fashioned idea of what medicine is all about. OHRP seems to think that doctors somehow know, by their clairvoyant insight, endowed when they pass their final exams, the best course of action for their patient, and they should go ahead and treat according to their ESP. I’d rather treat according to my EBM.

Good modern science based medicine is all about making a decision with your patient (or their family) based on their particular situation, the evidence base which exists, their values and desires, and the final calculation of risk and benefit. ‘In the face of uncertainty’ means that there is no clear answer. So, for example, I have a 3 day old baby at 25 weeks with a large PDA, who is still intubated, and no current clear signs of hemodynamic compromise. The OHRP thinks that I should decide whether or not to treat based on my view of what is best. Very often (as in this situation) my view is that I don’t know what is best, that is the only honest evaluation of that situation. If I don’t know what is best, then I can guess, use prejudice, treat according to what happened the last time, phone a friend, check what day of the week it is, or toss a coin.

Which of those satisfies my obligation to my patient?

They all do to some extent, because I am uncertain. The most ethical approach is the last one. To toss a coin. And to do so in the setting of a prospective experiment, where the data will be collected, the comparison of the different approaches can be made objectively, and the results made available to colleagues around the world. That fully satisfies my ethical obligation to do the best for my individual patient, and also means that sometime in the future I will have the knowledge to make a better decision, that there will be less uncertainty.

Also, at least if I toss the coin, if in the future it turns out that one of the 2 courses of action, giving ibuprofen, or not giving ibuprofen, is proven to be preferable, then at least 50% of my babies will have had the right treatment.  That is better than finding out that maybe 100% of them got the worse treatment!

To suppose that researchers are somehow working to a different ethical standard is entirely mistaken. My first responsibility is to my patient, and if I am reasonably certain that one arm of a comparative trial is not in their best interest, then I am morally obliged to not ‘toss the coin’ but to treat according to that best interest.

There are many ethicists, clinicians and clinical researchers who ‘get it’, the signatories of the letter in the PNEJM for example, why don’t the OHRP?

Posted in Neonatal Research | Leave a comment