I think my first training in resuscitation began with the principles outlined in the NRP 3rd edition program. As we have moved through subsequent editions with the current edition being number 7, I can’t help but think about how many changes have occurred over that time. One such change has been the approach to using medications as part of a resuscitation. Gone are such things as calcium gluconate, naloxone and sodium bicarbonate but something that has stood the test of time is epinephrine. The dosing and recommendations for administering epinephrine have changed over time as well with the dose of endotracheal medication increasing from 0.01 to 0.03 and now to 0.05 – 0.1 mg/kg. While this dosing has increased, that of IV administration has remained the same at 0.01 to 0.03 mg/kg. The change in dosing for the ETT route was due to an increasing awareness that this route just isn’t as effective as IV. Having said that with only 0.1% of resuscitations requiring such support the experience with either route is fairly limited.
What is the concern?
Giving a medication directly via the IV route ensures the dose reaches the heart in the amount desired. In the case of ETT administration there are a few potential issues along the way. The first is that one needs to push the dose down the ETT and this presumes the ETT is actually in the trachea (could have become dislodged). Secondly, if the medication is sent to the lung what effect does the liquid component in the airways have in terms of dilution and distribution of the medication? Lastly, even if you get the epinephrine to the lung it must be picked up at the capillary level and then returned to the left side of the heart. In the absence of significant forward pulmonary blood flow this is not assured.
What is the evidence?
In terms of human clinical research it remains fairly limited. Barber published a retrospective review of 47 newborns who received epinephrine via the endotracheal route. The study Use and efficacy of endotracheal versus intravenous epinephrine during neonatal cardiopulmonary resuscitation in the delivery room found that spontaneous circulation was restored in 32% of this cohort. Following the first dose, a subsequent dose of intravenous epinephrine restored circulation in 77%. This study provided the first suggestion that the IV route may be better than endotracheal. Keep in mind though that this study was retrospective and as the authors conclude in the end, prospective studies are needed to confirm these findings. The question really is what is the likelihood of restoring circulation if the first dose is given IV?
Eleven years later we have a second study that attempts to answer this question although once again it is retrospective. Efficacy of Intravenous and Endotracheal Epinephrine during NeonatalCardiopulmonary Resuscitation in the Delivery Room by Halling et al. This study really was designed to answer two questions. The study group looked at the period from July 2006 to July 2014. During this period the dose of IV epinephrine remained unchanged as per NRP recommendations but the dose of endotracheal epinephrine increased from 0.01 to 0.03 and then to 0.05 mg/kg endotracheally. The increase was in response to both NRP and site observations that the lower doses were not achieving the effect they were hoping for.
|Return of circulation
In the ETT group all doses except for 3 after the first dose were given as IV. There was no difference in the response rate over time suggesting that higher doses do not truly increase the chance of a better response. The authors noted that the effectiveness of the two arms were not that different despite a significantly higher dose of epinephrine being administered to the group receiving ETT epinephrine first which is not surprising given the higher recommended dosages.
What I find interesting though is that giving the first dose of epinephrine was given IV in 20 of the paitents, if it is indeed the better route one might expect a better response than in the ETT group. The response from one dose of ETT epi was 20% while that from the IV first group was in fact also only 20%! We do indeed need to be careful here with small numbers but the results at least to me do not suggest strongly that giving IV epi first ensures success. What the study suggests to me is that two doses of epinephrine may be needed to restore circulation. If you choose to start with IV it certainly does not seem unwise but if you have any delays I don’t see any reason to avoid ETT epinephrine as your first line.
The reality is that for many individuals a UVC is a procedure that while they may have learned in an NRP class they may have never actually placed one. Having an ETT in place though seems like a good place to start. I doubt we will ever see a randomized trial of ETT vs IV epinephrine in Neonatology at this point given the stance by the NRP so these sorts of studies I suspect will be the best we get.
For now, based on what is out there I suggest use the route that you can get first but expect to need additional doses at least one more time to achieve success. Lastly remember that even if you do everything correct there will be some that cannot be brought back. Rest assured though that if the first dose was given via ETT you have still done your best if that was the route you had.
The other day I met with some colleagues from Obstetrics and other members from Neonatology to look at a new way of configuring our delivery suites. The question on the table was which deliveries which were always the domain of the high risk labour floor could be safely done in a lower acuity area. From a delivery standpoint they would have all the tools they need but issues might arise from a resuscitation point of view if more advanced resuscitation was needed. Would you have enough space for a full team, would all the equipment you need be available and overall what is in the best interests of the baby and family?
We looked at a longstanding list of conditions both antenatal and intrapartum and one by one tried to decide whether all of these were high risk or if some were more moderate. Could one predict based on a condition how much resuscitation they might need? As we worked our way through the list there was much discussion but in the end we were left with expert opinion as there was really no data to go by. For example, when the topic of IUGR infants came up we pooled our collective experience and all agreed that most of the time these babies seem to go quite well. After a few shoulder shrugs we were left feeling good about our decision to allow them to deliver in the new area. Now several days later I have some concern that our thinking was a little too simple. You see, conditions such as IUGR may present as the only risk factor for an adverse outcome but what if they also present with meconium or the need for a instrument assisted delivery. We would presume the risk for advanced resuscitation (meaning intubation or chest compressions and/or medication need) would be increased but is there a better way of predicting the extent of this risk?
Indeed there might just be
An interesting approach to answer this question has been taken by an Argentinian group in their paper Risk factors for advanced resuscitation in term and near-term infants: a case–control study. They chose to use a prospective case control study matching one case to 4 control infants who did not require resuscitation. The inclusion criteria were fairly straightforward. All babies had to be 34 weeks gestational age or greater and free of congenital malformations. By performing the study in 16 centres they were able to amass 61953 deliveries and for each case they found (N=196) they found 784 deliveries that were matched by day of birth. The idea here was that by matching consecutive patients who did not require resuscitation you were standardizing the teams that were present at delivery.
The antepartum and intrapartum risk factors that were then examined to determine strengths of association with need for resuscitation were obtained from the list of risks as per the NRP recommendations.
A Tool For All of Us?
What came out of their study was a simple yet effective tool that can help to predict the likelihood of a baby needing resuscitation when all factors are taken into account. By resuscitation the authors defined this as intubation, chest compressions or medications. This is pretty advanced resuscitation! In essence this is a tool that could help us answer the questions above with far better estimation than a shoulder shrug and an “I think so” response. The table can be found by clicking on this link to download but the table looks like this.
By inserting checks into the applicable boxes you get a calculated expected need for resuscitation. Let’s look at the example that I outlined at the start of the discussion which was an IUGR infant. It turns out that IUGR itself increases the background risk for infants 34 weeks and above from 6% to 55% with that one factor alone. Add in the presence of fetal bradycardia that is so often seen with each contraction in these babies and the risk increases to 97%! Based on these numbers I would be hesitant to say that most of these kids should do well. The majority in fact would seem to need some help to transition into this world.
Some words of caution
The definition here of resuscitation was intubation, chest compressions or medications. I would like to presume that the practioners in these centres were using NRP so with respect to chest compressions and medication use I would think this should be comparable to a centre such as ours. What I don’t know for sure is how quickly these centres move to intubate. NRP has always been fairly clear that infants may be intubated at several time points during a resuscitation although recent changes to NRP have put more emphasis on the use of CPAP to establish FRC and avoid intubation. Having said that this study took place from 2011 – 2013 so earlier than the push for CPAP began. I have to wonder what the effect of having an earlier approach to intubating might have had on these results. I can only speculate but perhaps it is irrelevant to some degree as even if in many cases these babies did not need intubation now they still would have likely needed CPAP. The need for any respiratory support adds a respiratory therapist into the mix which in a crowded space with the additional equipment needed makes a small room even smaller. Therefore while I may question the threshold to intubate I suspect these results are fairly applicable in at least picking out the likelihood of needing a Neonatal team in attendance.
Moreover I think we might have a quick tool on our hands for our Obstetrical colleagues to triage which deliveries they should really have us at. A tool that estimates the risk may be better than a shoulder shrug even if it overestimates when the goal is to ensure safety.
After the recent CPS meeting I had a chance to meet with an Obstetrical colleague and old friend in Nova Scotia. It is easy to get lost in the beauty of the surroundings which we did. Hard to think about Neonatology when visits to places like Peggy’s Cove are possible. Given out mutual interest though in newborns our our conversation eventually meandered along the subject of the new NRP. What impact would the new recommendations with respect to meconium have on the requirements for providers at a delivery. This question gave me reason to pause as I work in a level III centre and with that lens tend to have a very different perspective than those who work in level I and II centres (I know we don’t label them as such anymore but for many of you that has some meaning). Every delivery that is deemed high risk in our tertiary centre has ready access to those who can intubate so the changes in recommendations don’t really affect our staffing to any great degree. What if you are in a centre where the Pediatrician needs to be called in from home? Do you still have to call in people to prepare for a pending delivery of a baby through meconium stained fluid?
What does the new recommendation actually say?
These recommendations are from the American Heart Association and are being adopted by the NRP committees in the US and Canada. The roll-out for this change is coming this fall with all courses required to teach the new requirements as of September 2017.
“However, if the infant born through meconium-stained amniotic fluid presents with poor muscle tone and inadequate breathing efforts, the initial steps of resuscitation should be completed under the radiant warmer. PPV should be initiated if the infant is not breathing or the heart rate is less than 100/min after the initial steps are completed. Routine intubation for tracheal suction in this setting is not suggested, because there is insufficient evidence to continue recommending this practice. (Class IIb, LOE C-LD)“
The rationale for the change is that is that there is a lack of evidence to demonstrate that routine suctioning will reduce the incidence of meconium aspiration syndrome and its consequences. Rather priority is placed on the establishment of adequate FRC and ventilation thereby placing a priority on teaching of proper bag-valve mask or t-piece resuscitator. Better to establish ventilation than delay while atempting to intubate and run the risk of further hypoxia and hypercarbia causing pulmonary hypertension.
Does this mean you don’t need to have a person skilled in intubation at such deliveries?
This question is the real reason for the post. At least from my standpoint the answer is that you do in fact still require such people. This may seem to be in conflict with the new position but if you move past that recommendation above you will see there is another line that follows afterwards that is the basis for my argument.
“Appropriate intervention to support ventilation and oxygenation should be initiated as indicated for each individual infant. This may include intubation and suction if the airway is obstructed.”
While we should not routinely perform such intubations there may be a time and a place. If one has intiated PPV with a mask and is not obtaining a rising heart rate, MRSOPA should be followed and attempts made to optimize ventilation. What if that is unsuccessful though and heart rate continues to be poor. You could have a plug of meconium distal to the vocal cords and this is the reason that intubation should be considered. In order to remove such a plug one would need to have an intubator present.
Where do we go from here?
As much as I would like to tell my colleague that he doesn’t need to have this skill set at a delivery for meconium I am afraid the skill still needs to be present. It will be interesting to see how instructors roll this out and answer such questions. It is a little concerning to me that in our world of wanting the “skinny” or “Coles’ Notes” version of things, the possibility of still needing the intubator on short notice may be lost. Having someone on call who is only “5 minutes away” may seem to be alright but at 3 AM I assure you the 5 minutes will become 15 as the person is woken, dresses, gets to the car and parks. Whether it is 5 or 15 minutes each centre needs to ask themselves if the baby is in need of urgent intubation are you willing to wait that amount of time for that to happen?
I hope not.
Ask almost anyone who has worked in the field of Neonatology for some time and they will tell you that babies are not as sick as they once were. We can give a lot of credit to better antenatal steroid use, maternal nutrition and general management during pregnancy. Additionally, after birth we now rush to place infants on CPAP and achieve adequate expansion of the lungs which in many cases staves off intubation.
The downside to our success though is that the opportunities to provide positive pressure ventilation (PPV) and moreover intubation are becoming less and less. How then do we perform when we are asked to do such procedures on an infrequent basis? The answer as you might expect is not that well. Dr. Schmolzer et al studied the ability of people to keep a good seal and found a 29% leak on average with as high as 63% in one patient. As this was a study in which people were being observed one might think the Hawthorne Effect might artificially decrease the percentage leak compared to real world scenarios when you know you aren’t being watched.
What is the cause of the leak?
Leaks most commonly occur on either side of the nasolabial folds. Although at least in my experience we educate trainees about this issue it remains a problem. I would also speculate that at the times when we need to be at our best during an advanced resuscitation involving chest compressions we may well function at our worst. This is the effect of the adrenalin rushing through our system as our sympathetic system turns into overdrive. The question therefore is one of getting around human error in particular when we need to minimize such inefficiency the most.
If the masks are prone to leaking and with it the ability to properly ventilate compromised, how could we minimize such human error. The answer may lie in what I consider to be an ingenious way to apply a mask. The concept and it is just that at the moment is to use suction to apply the mask to the face without risk of leak. Lorenz L et al have just published a proof of concept study utilizing a mannequin with a “seal skin” layer applied to the face to simulate human skin. The article is entitled A new suction mask to reduce leak during neonatal resuscitation: a manikin study.
In this study, the mask was applied to the face of the mannequin and 100 cm H2O pressure was applied through a side port on the mask. There is an inner and outer ring such that the internal 41 mm diameter mask is surrounded by a double wall in which the suction is applied to the space between the two walls leading to the mask seating itself firmly against the face. The authors then studied the amount of leak found when using a Neopuff set to deliver 40 – 60 breaths per minute at pressures of 25/5. For this study 60 courses were tested.
How did it do compared to PPV through a traditional mask?
As you might expect (since you can feel my excitement!) it did very well. The average leaks using a conventional approach were quite good at 12.1% but the suction mask was only 0.7% leak. Importantly the ranges were quite different. PPV through a conventional mask had a range of 0.6 – 39% leak while with the suction version it was 0.2 – 4.6%. These results were statistically different.
What does the future hold?
As mentioned this study is what one would consider a proof of concept study. We do not know how this would fare in the real world and that of course is the next step. In terms of harm, the authors did note that when applied to the forearm of an adult it caused some mild redness from the suction that vanished quickly on breaking of the seal but we do not know if there could be greater harm in a newborn in particular one who is quite small. Such testing will be needed as part of any further study.
Having said that I think this rethink of the mask for PPV could be transformative to those who perform neonatal resuscitation infrequently. If this mask is found not only to be effective in a clinical trial but safe as well I would suggest a change to this type of mask could quite literally be life saving. Placed in the hands of those who are inexperienced in keeping a seal, PPV would become much more effective and in particular for rural sites the infants being transported in much more stable than some are at present.
Keep your eyes peeled for future work using this mask. Something tells me if it proves to be efficacious outside of a seal skin covered mannequin, your toolkit for providing NRP may be in for a change.
As I read through the new NRP recommendations and began posting interesting points on my Facebook Page I came across a section which has left me a little uneasy. With respect to a newborn 36 weeks and above who is born asystolic and by ten minutes of age continues to remain so and has an apgar score of zero the recommendation that has been put forward is this:
An Apgar score of 0 at 10 minutes is a strong predictor of mortality and morbidity in late-preterm and term infants. We suggest that, in babies with an Apgar score of 0 after 10 minutes of resuscitation, if the heart rate remains undetectable, it may be reasonable to stop resuscitation; however, the decision to continue or discontinue resuscitative efforts should be individualized. Variables to be considered may include whether the resuscitation was considered to be optimal, availability of advanced neonatal care, such as therapeutic hypothermia, specific circumstances before delivery (eg, known timing of the insult), and wishes expressed by the family (weak recommendation, very-low-quality evidence).
There are some significant problems with this part of the statement.
- They claim that the apgar score at ten minutes is a strong predictor but when you look at the analysis of the evidence presented in the body of the paper it is weak at best. I am not clear how one declares the prediction is strong in the face of poor evidence but I will acknowledge intuitively that this makes some sense but do challenge them on the use of the word “strong”.
2. They are correct in acknowledging that the introduction of hypothermia in such settings has changed the landscape in as much as I find it quite difficult to prognosticate unless a child is truly moribund after resuscitation. Given such uncertainty it is concerning to me that this recommendation may be committed to memory incorrectly in some places that do have access to cooling and may be used more rigidly as though shalt stop at 10 minutes.
3. In the middle of a resuscitation it is quite difficult to process all of the facts pertaining to a particular newborn while orders for chest compressions, emergency UVCs and epinephrine are being given. Can we really individualize within ten minutes accurately and take the families wishes truly into account? This just does not seem practical.
4. The families wishes are taken into account but inserted as a “weak recommendation”. How can the wishes of the family in any family centred model of care be minimized in such a way even if we believe the situation to be dire?
5. Since the introduction of hypothermia there appears to be a near 50% survival rate in such newborns and as the authors state 27% of survivors who received cooling had no moderate or severe disability. Here in lies my greatest issue with this guideline and that is the hypocrisy this position takes when you compare populations at 23 and 24 weeks gestational age. Survival at these GA in the recent NEJM study of almost 5000 preterm infants under 27 weeks were 33 and 57 % respectively at 23 & 24 weeks with rates of survival without moderate or severe disability being 16 and 31% in the two groups. The fallout from this and other studies at the extremes of gestational age have been that we should be more aggressive as the outcomes are not as bad as one would predict. How can we argue this for the 23-24 week infants and for term infant with the same likelihood of outcomes we would unilaterally stop in many centres?!
So Now What Do We Do?
We are supposed to be practising family centred care and much like the argument at the edge of viability the same should apply here. The wishes of the family should never be minimized. Arguably it may be very difficult in such an unexpected scenario to appraise a family of the situation and have clarity around the issue but if a heart rate can be restored after a few more minutes do we not owe it to the family and the child to bring the infant back to the NICU and see what transpires especially if cooling is available?
The million dollar question of course is where do we draw the line? No heart rate at 15, 20 minutes? Based on the evidence thus far it seems to me that a little longer than 10 minutes is reasonable especially in well equipped centres with access to cooling and modern ventilation and treatments for pulmonary hypertension. How long though must be individualized and should be determined in partnership with the team caring for the patient which must include the family.