Timing of MRI after HIE

Three recent articles have investigated whether we should wait until a week or so after birth to perform brain imaging in infants with encephalopathy, or whether earlier imaging might be just as predictive.

The three articles have consistent findings, which is remarkable in itself! All three note that infants at high risk, most of whom have undergone hypothermia treatment, when they have MRI at 2 to 4 days of age, the results are very similar to the findings if you wait until a week or so to do the study.

Agut T, et al. Early identification of brain injury in infants with hypoxic ischemic encephalopathy at high risk for severe impairments: accuracy of MRI performed in the first days of life. BMC Pediatrics. 2014;14(1):177.

Boudes E, et al. MRI obtained during versus after hypothermia in asphyxiated newborns. Archives of Disease in Childhood – Fetal and Neonatal Edition. 2015;100(3):F238-F42.

Skranes JH, et al. Brain imaging in cooled encephalopatic neonates does not differ between four and 11 days after birth. Acta Paediatrica. 2015.

Now I think you could say the same about many of these type of studies as about the studies of brain imaging in very preterm babies. Which is, that if the reason that you want to do the study is to aid in medical decision making (which is explicitly stated in the first of the 3 articles) we need much better data of the positive predictive value of the findings for profoundly adverse outcomes.

The best data I think come from the analysis of the TOBY trial, Rutherford M, et al. Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy: a nested substudy of a randomised controlled trial. The Lancet Neurology. 2010;9(1):39-45. That study used a scoring system, created by the same group, and showed that the positive predictive value of moderate or severe lesions in the basal ganglia and thalami, severe white matter lesions, or an abnormal posterior limb of the internal capsule for death or severe disability at 18 months of age was 0·76 (95% CI 0·65–0·87).

Severe disability was defined as at least one of: mental development index (MDI) less than 70 (2 or more SD below the mean) on the Bayley infant scales (BSID II) at 2 years;  cerebral palsy with a GMFCS of 3–5 (unlikely to be ambulant) or bilateral cortical visual impairment with no useful vision.

There is some evidence that a 2 year Bayley is more predictive of limited longer term functioning after HIE than it is for former extreme preterm infants, for example this article from the follow up of the NRN trial, Pappas A, et al. Cognitive outcomes after neonatal encephalopathy. Pediatrics. 2015;135(3):e624-34. Of 30 babies with a Bayley 2 MDI less than 70, 27 of them had a full scale IQ less than 70 at 6 to 7 years of age. (Most of the infants with an MDI less than 70 were below 55, 24 of the 30; also 23 of the 31 babies with an IQ below 70 were below 55).

If we put all this together it seems that it might be possible to have a reasonably accurate prediction of severely abnormal outcome using MRI shortly after, or even during the final day of, therapeutic hypothermia.  I think before we rush to performing early MRI, and use them for decision making, we should have more, and more direct, evidence that a certain severity of abnormality on the early MRI, accurately predicts profound impairment, and that this is better than clinical examination, or other predictive indices.


Posted in Neonatal Research | Tagged , , , , | Leave a comment

Bubble Up

In recent years the introduction of Helping Babies Breathe in low-income countries has proven to be effective in reducing fresh stillbirth rates. Babies who would otherwise be considered to be fresh stillbirths are given a chance, helped to breathe and often survive without any complications. Vossius C, et al. Cost-Effectiveness of the “Helping Babies Breathe” Program in a Missionary Hospital in Rural Tanzania. PLoS ONE. 2014;9(7):e102080. This study for example showed a cost per life saved of just over 200 dollars, over the first year of the program. As the retraining is done without new equipment and during the hours of work, it costs next to nothing, so the costs per life saved should drop even further as the years go by.

Some babies do go on to need respiratory support, which is itself costly and in limited supply in middle resource countries.

A new paper from Nicaragua describes the results of introducing and promoting the use of bubble nasal CPAP in an NICU that was already providing assisted ventilation. (Rezzonico R, et al. Impact of the systematic introduction of low-cost bubble nasal CPAP in a NICU of a developing country: a prospective pre- and post-intervention study. BMC Pediatrics. 2015;15(1).) They showed that after the change in practice there were many more babies who only required CPAP, the proportion more than doubled to over 60%, and the mortality rate decreased despite an increase in admissions.

Posted in Neonatal Research | Tagged , , | Leave a comment

Not Neonatology : Photos from the antipodes

I have re-organized the photos on the photograph pages (click on the menu under the header banner), so each gallery now occupies a separate page. At the same time I have added and re-organized a page on Australian wildlife without wings, and a page on New Zealand birds, some of which have appeared before on other posts, and some are newly posted.

I hope you enjoy them.

Posted in Neonatal Research | 1 Comment

A Facebook Page

Sainte Justine Neonatology now has a facebook page.

You can find it here : http://www.facebook.com/neonatologie

Posted in Neonatal Research | 2 Comments

Off-label NO: No… or Yes?

Recent articles about off-label use of inhaled Nitric Oxide (iNO) in the preterm infant shows that such use is :

a. extremely variable from one center to another, (Truog WE, et al. Inhaled nitric oxide usage in preterm infants in the NICHD neonatal research network: inter-site variation and propensity evaluation. J Perinatol. 2014.)

and b. increasing in some places.  Ellsworth MA, et al. Off-Label Use of Inhaled Nitric Oxide After Release of NIH Consensus Statement. Pediatrics. 2015;135(4):643-8.)

As you can tell from the title of the 2nd article, the authors wished to see what effect the release of the NIH consensus statement had. I was involved in the hearings that led to production of that statement, which concluded, as the authors of this article put it :

The resulting NIH statement and American Academy of Pediatrics clinical report indicated that available evidence did not support the routine use of iNO in preterm infants, though rare clinical situations and new management strategies did merit further study.

The first of these 2 references suggested that use might have been decreasing after the consensus statement, at least among the extremely low birth weight infants in the Neonatal Research Network generic database. The second article suggests the opposite, in the NICUs in the Pediatrix network. In those units iNO is used more often off-label than ‘on-label’, very nearly 50% of all the NO days are in babies less than 29 weeks gestation, and those babies average 7 days of treatment, compared to 5 days for the term babies.

If you recall, both the Cochrane Systematic Review (Barrington KJ, Finer N. Inhaled nitric oxide for respiratory failure in preterm infants. Cochrane database of systematic reviews (Online). 2010;12:CD000509), and the IPD meta-analysis that we published (Askie LM, et al. Inhaled Nitric Oxide in Preterm Infants: An Individual-Patient Data Meta-analysis of Randomized Trials. Pediatrics. 2011;128(4):729-39.) show no clear advantage of iNO in any subgroup of preterm babies. The possible benefit in babies with early evolving chronic lung injury, shown in the NOCLD trial, was not confirmed by the NewNO trial, presented at a Hot Topics, and not yet published, (As soon as it is published we will be updating the Cochrane Review).

So why is there still any use in the preterm? Are we all anti-evidence based medicine cooks?  I don’t think so, I think there are still unanswered questions in the preterm, and as a result situations where it is reasonable to consider the use of iNO.

The studies of early rescue use of iNO, which show no benefit (and a possible increase in intracranial hemorrhage) enrolled preterm infants with severe respiratory failure. But only a minority of those infants have severe pulmonary hypertension, in particular those who had oligohydramnios after PPROM. Some babies like that have a very limited response to surfactant, and after surfactant they have clear lungs on x-ray, and often have a big ductal saturation gradient, i.e. evidence of right to left shunting and supra-systemic PA pressures. When I have given iNO to such babies (and others have reported this also) they often have a dramatic response and go from being hypoxic to rapid weaning of their oxygen requirements.

When we did the IPD meta-analysis of iNO in the preterm we wanted to specifically address whether preterm infants with evidence of PPHN had a different response, and potentially a benefit from iNO. Unfortunately there were very few babies who had a record of either having, or not having, pulmonary hypertension. So although there were more survivals without BPD among preterm infants who had a diagnosis of PPHN it was not clear whether or not this might have been due to chance. Other case series, and one subgroup analysis of one of the RCTs, also suggest that there may be a benefit, but all we can say at present is that such babies often have a big improvement in oxygenation in the short term.

The other situation in which iNO is being used is during acute deteriorations in oxygenation in babies who are developing BPD. In those babies you can also get improvements in oxygenation, even among those who don’t have clear evidence of pulmonary hypertension, presumably because you improve VQ matching.

I think, therefore that there is room to perform more trials in those 2 groups of babies, to see whether the acute improvements in oxygenation lead to improved survival or reduced complications.

Randomizing very sick, hypoxic, preterm babies who have clear evidence of PPHN in the first few hours of life to iNO or placebo will be difficult, but not impossible. It could even be done with a rescue phase for babies who stay hypoxic after a period of placebo use. It is possible that the placebo babies would gradually improve without iNO, and that improving gradually, rather than suddenly, would lead to a smoother clinical course, and less intracranial hemorrhage.

The second group of babies would be easier to study, I think; babies who at a couple of weeks of age have signs of early lung injury, and very high oxygen requirements, this represents about half of the babies in the Ellsworth publication. In those babies, who often get iNO for very long periods if there is an initial response, I think there is a chance of benefit, but a real risk of very high costs without overall clinical benefit.

This is not the same group of babies as studied in NOCLD or the NewNO trial. The babies in those trials were typically on low oxygen and respiratory support, so it will require targeted trials to investigate the highest risk babies.

Now I know I’m on the right track when what I say agrees, largely, with what Neil Finer and Nick Evans write! (Finer NN, Evans N. Inhaled Nitric Oxide for the Preterm Infant: Evidence Versus Practice. Pediatrics. 2015;135(4):754-6).

Posted in Neonatal Research | Tagged , , | 1 Comment

Bevacizumab or that other more expensive one?

A very interesting and depressing group of articles in today’s BMJ. Retinopathy of prematurity is a small market compared to wet macular degeneration in the elderly, in whom VEGF inhibitors are proven to be extremely effective. Ranibizumab (I will call it RBZ) is far more expensive than BVZ for the effective dose, and the company responsible refuses to apply for a licence for BVZ for this indication, despite the 6 comparative trials which show equal efficacy, and a Cochrane review showing equal toxicity. The companies’ tactics are appalling:

The NHS spends £244m a year on ranibizumab, the second highest amount for any drug. However, research and development costs do not explain why ranibizumab is priced 10-20 times higher than bevacizumab, Philip Rosenfeld, professor of ophthalmology at the Bascom Palmer Eye Institute in Florida told The BMJ. He was involved in the early phase trials of ranibizumab and was one of those who pioneered the use of bevacizumab for wet AMD. He said bevacizumab is also more expensive to make than ranibizumab.

Ranibizumab is a monoclonal antibody fragment derived from the same parent monoclonal antibody as bevacizumab. Both drugs act by inhibiting vascular endothelial growth factor (VEGF), preventing blood vessel growth. Roche holds the intellectual property rights for both, although Novartis has the rights to market ranibizumab in Europe. Bevacizumab is licensed for use only in cancer conditions, and Roche has never applied for a marketing authorisation for ophthalmic conditions—despite repeated calls from politicians to do so. This means only the more expensive ranibizumab has such a licence, and prescriptions of bevacizumab— even in its repackaged form for ophthalmic use—are “off label,” …..

But eight years on, bevacizumab is scarcely used for AMD in the NHS, and the companies have not done the necessary trials. ……  Minutes from a NICE workshop show that Roche said decisions not to develop bevacizumab for ocular use have been “due to corporate considerations.”

The article goes on to say :

An accompanying article shows how the manufacturers didn’t want to do the trials themselves and when it was agreed that the public should fund both a comparative and a dosing trial, they did all they could to scupper them—even turning to the Royal National Institute of Blind People (RNIB) for help. Then once these trials were published, they embarked on a campaign to undermine and divert attention from the results, raising safety concerns themselves and via their key opinion leaders and charities.

That accompanying article describes the way Novartis has tried to wreck other trials.

The need to compare bevacizumab and ranibizumab for ophthalmic use was perceived to be a priority for the NHS. And so the “randomised controlled trial of alternative treatments to inhibit VEGF in age-related choroidal neovascularisation” (IVAN) trial was conceived, at a cost to the public of about £10m……

However, there was resistance to the trial from the outset. Barney Reeves, professorial research fellow in health services research at Bristol University and one of the IVAN trialists, told The BMJ that it was a difficult trial to do politically. “The drug manufacturers didn’t want it done,” he said, adding: “Novartis tried to prevent UK ophthalmologists joining the IVAN trial, with their sales representatives lobbying potential principal investigators against the trial and telling them that the IVAN protocol was seriously flawed.”

Emails obtained by The BMJ under a freedom of information request show that clinicians with ties to Novartis urged some primary care trusts to pull out of the trial. The emails show discussions between ophthalmologists close to Novartis about how “big centres” in the US pulled out of the CATT trial because they would lose industry funding for other trials. The emails said that this situation “could be applied to the UK”—that is, researchers would be put off IVAN for fear of losing industry studies.

Emails also show that Novartis approached ophthalmology centres to conduct trials into other eye conditions for them at the same time IVAN was progressing. Novartis helped the investigators in these industry funded trials to obtain supplies of ranibizumab, in stark contrast to its approach to the IVAN trialists.

A new trial, not included in those systematic reviews has just been published in the NEJM. It showed that in diabetic macular oedema, nine shots of BVZ ($50 per dose), RBZ ($1,200 per dose), and aflibercept ($1,950) produced an equivalent benefit.

It was accompanied by an editorial which ended thus:

“We believe that all financial incentives and logistic barriers to providing the least expensive drug, among drugs equivalent in safety and efficacy, should be eliminated so that patients may benefit fully from the results of this Diabetic Retinopathy Clinical Research Network trial as well as those from other comparative trials.”

Which sounds eminently sensible, but not likely to hold much sway with Novartis, whose main consideration appears to be their own benefit, and not the patients.

Extreme preterms are small fry of course, in many ways. A single treatment, in a small number of babies, is not going to make a huge amount of money for anyone, unless you can find a way to restrict the treatment choices to only the more expensive drug. So Novartis are sponsoring a trial of RBZ against laser (NCT02375971), when, if we need to study RBZ at all, what is needed is a trial of RBZ against BVZ. Unfortunately such a trial would be likely to show that they are equivalent, as they are in adults, so I don’t hold out much hope that Novartis will fund such a trial.


Posted in Neonatal Research | Tagged | 1 Comment

What’s new in non-invasive respiratory support?

A round-up of a few publications that have appeared over the last few months.

Lampland AL et al: Bi-level CPAP does not improve gas exchange when compared with conventional CPAP for the treatment of neonates recovering from respiratory distress syndrome. Archives of Disease in Childhood – Fetal and Neonatal Edition 2015, 100(1):F31-F34.  In this study, 20 LBW babies were observed in a cross-over design. They were randomized to either get CPAP at +6, or SiPAP with a mean of 6 (it was usually about 9/4 with 20 1 second cycles per minute), they used the Infant Flow system, and all the babies had a chin strap in place. the main outcome variables were oxygenation and transcutaneous CO2. Treatment blocks were for 1 hour at a time. There were basically no differences between groups in gas exchange, and also no difference in apnea. But I question whether 1 hour is really enough time to see if there is an effect on apnea, the mean number of prolonged apneas was less than 1 per hour. Apnea occurs in unpredictable patterns, so much longer treatment periods are needed to see if there is an effect.

Gizzi C et al: Is synchronised NIPPV more effective than NIPPV and NCPAP in treating apnoea of prematurity (AOP)? A randomised cross-over trial. Archives of Disease in Childhood – Fetal and Neonatal Edition 2015, 100(1):F17-F23. This is also a randomized cross-over study, with, this time, 4 hour periods on each of 3 modes of ventilation. CPAP at 5 to 6, Nasal ventilation using the same device (Giulia ventilator) and set at 15/5 to 6, rate of 20 and Ti of 0.3s. Synchronised nasal ventilation was achieved by adding a pneumotach to the circuit, between the Y piece and the patient prongs. They used tight fitting prongs, and, during the SNIPPV sessions one of the investigators sat by the bedside to make sure the flow signals and triggering were working properly. They showed a reduction in desaturation spells with SNIPPV compared to non-synchronised NIPPV and compared to CPAP, with a p-value of very close to 0.05 for both comparisons.  There was no change in bradycardias, which were uncommon in any case. I really don’t think that it would be possible with flow triggering to have prolonged synchronised NIPPV, the leaks are too variable, so synchronisation without an investigator sat by the bedside to constantly adjust the prongs is unlikely, I think, to work.

Stern DJ et al: Synchronized neonatal non-invasive ventilation-a pilot study: The graseby capsule with bi-level NCPAP. Pediatric Pulmonology 2014, 49(7):659-664. If you have the Infant Flow SiPAP device you can try to synchronize to the babies effort using the Graseby Capsule. But does it actually work? In these authors hands, yes. As long as you have the Ti set to at least 0.3 seconds, and you don’t expect it to trigger 100% of the time. They got about 3/4 of the respiratory efforts to trigger an inhalation, and no triggers when the baby did not breathe. It took about 26 ms to respond, which was actually faster than the respiratory inductance device. I think with care, you could synchronize for prolonged periods with the Graseby capsule, it used to work with the Infantstar ventilator. It was that system I used for my RCT of post-extubation NIPPV compared to CPAP. Incidentally the RCTs showing benefit (in terms of reduction of extubation failure with NIPPV) mostly used synchronisation, mostly with the Infantstar ventilator and a Graseby capsule.
Salvo V et al: Noninvasive Ventilation Strategies for Early Treatment of RDS in Preterm Infants: An RCT. Pediatrics 2015, 135(3):444-451. In this study 124 VLBW infants who did not need intubation in the delivery room were randomized to either nasal synchronized IPPV using the same system as in the Gizzi study above, or to BiPAP using the Infant Flow device. On BiPAP the babies got 8 to 9 cmH2O and a PEEP of 4 to 6, with an time on the high pressure of 1 second and a rate of 20 per minute. on nSIPPV they got 15 to 20 over 4 to 6, with a frequency of 40 and a time on  the high pressure of 0.3 to 0.4s.

The babies of 26 weeks and less were treated with surfactant via a brief intubation first, the more mature babies didn’t routinely get surfactant.

Failure of non-invasive support was defined as needing more than 40%, or having high CO2 (over 65) or lots of apnea. There were the same number of babies who filed in each group. (Some of the failures were for reasons that are not included in their definition of failure, such as pulmonary hypertension, PDA and pneumothorax).

Nevertheless, they had no differences in any outcomes.

One could again question how successful the synchronisation was over several days, and also how well maintained the pressures were on the BiPAP, infant flow device, which often delivers pressures much less than those that you want. Nevertheless in routine clinical practice this study doesn’t suggest any large difference between the efficacy of these two approaches.

Shi Y et al: A prospective, randomized, controlled study of NIPPV versus nCPAP in preterm and term infants with respiratory distress syndrome. Pediatric Pulmonology 2014, 49(7):673-678.

I started to write about this trial, but the more I looked at it the more uninterpretable it became. This is reported as an RCT in term and preterm babies with respiratory distress who were randomized to NIPPV or CPAP. About 90 in each arm of the trial. One of the outcome variables (the primary outcome) is need for intubation, and apparently 88% of the NIPPV babies avoided intubation, and 79% of the CPAP babies. Which is not significant by Chi-square, p=0.13 (with or without Yates correction). Although the authors state that it was significant, p<0.05.

Also 83% of the babies in each group received surfactant. I am not sure how. as they weren’t intubated! So I think the data are unreliable, unless some clarification can be obtained. I don’t understand how the reviewers didn’t pick this up.

Posted in Neonatal Research | Tagged , | Leave a comment