With American Thanksgiving coming up this weekend a post about “cold turkey” seemed apropos. You can’t work in Neonatology and not be familiar with CPAP. We have learned much about this modality in the last couple decades as clinicians have moved more and more towards non-invasive support as the preferred strategy for supporting newborns regardless of gestational age. Ask a Neonatologist how they use CPAP and you will find varied opinions about how high to go and how quickly to wean. I have written about one weaning strategy before on this blog using monitor oxygen saturation histogram data to make such decisions Improve your success rate in weaning from CPAP. One question though that has often been asked is what level of CPAP is best to remove a baby from? In particular for our smallest infants who may have BPD or reduced pulmonary reserve due to lower numbers of alveoli as they continue to develop should you discontinue at +5, +4 or +3? This question is what some creative authors from Texas sought to answer in the paper being discussed today.
To Wean or Not To Wean?
Kakkilaya V et al published Discontinuing Nasal Continuous Positive Airway Pressure in Infants ≤32 Weeks Gestational Age: A Randomized Control Trial in the Journal of Pediatrics this October. The authors studied infants from 23+0 to 32+6 weeks gestational age at birth and looked at whether a strategy of discontinuing from +5 or weaning from +5 to +3 then stopping resulted in fewer failures from stoppage. Infants were recruited in two ways. Some infants were intubated with planned extubation to pressures from +5 to +8 while others were on CPAP always. The study included 226 infants or which 116 were assigned to control so had removal of CPAP at +5 if after 24 hours they met the stability criteria below. The other 110 infants reduced CPAP from +5 once every 24 hours if the same criteria were met. Reasons for restarting CPAP were also as shown below at the bottom of Table 1. If an infant failed then they went back to the level of CPAP they had been on previously when stability criteria were met. Once they had stability criteria at that level again for 24 hours the wean could resume.
Did they manage to find a difference?
Table 5 reveals the significant finding here which is that for the primary outcome there was no difference and it didn’t matter whether the infants were ventilated or not. One finding that was different was the number of neonates who failed to stop CPAP two or more times. This favoured the weaning approach. Aside from that the groups were comparable and there really wasn’t much benefit seen from one approach versus the other.
Thoughts About the Study
The study was a fairly straightforward one and although there wasn’t a significant result found there are some questions that I think we can think about.
The stability criteria did not have results from histogram analysis included as a measure of stability. I can’t help but wonder if addition of this approach would have identified some infants who were actually not ready to wean. Having said that, one challenge is to come to an agreement on what a stable histogram is. Based on a survey from my own colleagues recently I would say like many things in Neonatology, we are all over the map. If this study were to be repeated using histograms for decisions on weaning some sort of agreement would be needed on what qualifies as a stable histogram.
Our group has already tended to use +4 as the final weaning step for our ELBW and VLBW infants based on anecdotal experience that many of these kids if stopped at +5 will fail even when they seem to be stable. Repeating this study looking at weaning from +4 to +3 before stopping vs stopping at +4 could be interesting as well.
Finally, I do wonder if the wean was too fast to show a difference. It is not uncommon practice in the smallest infants to keep them on +4 for a couple days even if it seems that the histograms would indicate the baby is ready to stop CPAP. Perhaps a weaning strategy of allowing a minimum of q48h instead of q24h would have found different results?
I do think the authors explored a great question and I would be reluctant here to “throw the baby out with the bathwater”. There is something here but based on the methodology (which I don’t think is flawed per se) I think they just couldn’t prove what I suspect is true.
I have written about non-traditional methods of providing surfactant to newborns previously. The practice of intubating a preterm infant to administer surfactant and leaving the endotracheal tube in with a slow wean of ventilation is mostly a thing of the past (at least in my units). Strategies have evolved and have seen the development of the INSURE technique, LISA methods, use of an LMA to delivery surfactant and even simple deposition into the pharynx all with variable success.
Poractant alfa at 200 mg/kg was used in this study and delivered via aerosolization using a vibrating membrane called the eFlow. The authors chose to look at infants from 29 0/7 to 33 6/7 weeks at birth and stratified them into two groups of 29 0/7 to 31 6/7 and 32 0/7 to 33 6/7 weeks. They estimated a need for 70 babies based on an anticipated failure rate of 30% in the control group vs 5% in the treatment group. Unfortunately, due to several reasons the study was only able to recruit 64 babies for randomization before being stopped due to the recruitment issues. The design of the study included adequate blinding with a sham procedure and there were predefined “failure criteria” necessitating intubation at the outset of the study. These criteria are acceptable to me as they are similar enough to my own practice and were:
1. FiO2 >0.35 over more than 30 min OR FiO2 >0.45 at
2. More than four apnoeas/hour OR two apnoeas requiring bag
and mask ventilation.
3. Two capillary blood gas samples with a pH <7.2 and partial pressure of carbon dioxide >65 mm Hg (or partial pressure
of carbon dioxide in arterial blood (PaCO2) >60 mm Hg if
arterial blood gas sample).
4. Intubation deemed necessary by the attending physician.
What did they find?
The primary outcome CPAP failure within 72 hours of birth was indeed different in the two groups.
CPAP failure by 72 hours
CPAP + surfactant
(RR (95% CI)=0.526 (0.292 to 0.950))
Clearly the event rates were quite off from what they expected in the power calculation but given that they found a difference as opposed to no difference at all the fact that they didn’t recruit the numbers they planned is of less importance.
However, what is interesting is when they looked at the planned analysis by stratification an interesting finding emerged.
Group 1 (29 0/7 to 31 6/7)
CPAP failure by 72 hours
CPAP + surfactant
(RR (95% CI)=0.860 (0.389 to 1.90))
Group 2 (32 0/7 to 33 6/7
CPAP failure by 72 hours
CPAP + surfactant
(RR (95% CI)=0.254 (0.089 to 0.727))
There were a number of secondary outcomes looked at as well which may be of interest to you but as the numbers here are quite small I will not comment other than to say there was no increased incidence of complications with surfactant administration in this fashion. Also for those who ultimately failed CPAP the time when they did so was quite delayed compared to CPAP alone. Age at intubation for nCPAP failure, hours 4.9 (2.7–10.6) 11.6 (9.0–31.1) 0.008*
What can we take from this?
I believe these results are encouraging even if the study is a small one. The message I take from this study is that aerosolization of surfactant delivers some amount of product to the lungs. Those with more significant RDS or smaller lungs (those in the 29 0/7 to 31 6/7 group) may not get enough surfactant to treat their RDS sufficiently to avoid intubation. Those with less significant RDS or a larger number of alveoli get “enough” of a dose delivered to the alveoli to make a difference and avoid intubation. It is worth stressing that there can be no specific comment about using this strategy in even more immature infants as they weren’t tested. If I had to guess though, I would expect no difference given the findings in the smaller group.
As a physician responsible for transport though I am interested in the potential benefits to those born in non-tertiary centres. Many centres lack individuals with the confidence and skill to regularly place endotracheal tubes. For these centres it may be that providing nebulized surfactant could delay the time to treatment failure, allowing more time for a trained transport team to arrive. Training of course would be needed in these centres on how to administer surfactant in this way but it is an interesting concept to consider. With a near tripling of the average time to treatment failure the extra hours on CPAP would be much appreciated when weather delays or difficulty securing air assets means long delays in transport team arrivals.
To be sure this isn’t the last study of this kind but it certainly is an interesting start and one that will no doubt produce questions that will help formulate the next study design.
A patient has been extubated to CPAP and is failing with increasing oxygen requirements or increasing apnea and bradycardia. In most cases an infant would be reintubated but is there another way? While CPAP has been around for some time to support our infants after extubation, a new method using high frequency nasal ventilation has arrived and just doesn’t want to go away. Depending on your viewpoint, maybe it should or maybe it is worth a closer look. I have written about the modality before in High Frequency Nasal Ventilation: What Are We Waiting For? While it remains a promising technology questions still remain as to whether it actually delivers as promised.
Better CO2 elimination?
For those who have used a high frequency oscillator, you would know that it does a marvelous job of removing CO2 from the lungs. If it does so well when using an endotracheal tube, why wouldn’t it do just as good a job when used in a non-invasive way? That is the hypothesis that a group of German Neonatologists put forth in their paper this month entitled Non-invasive high-frequency oscillatory ventilation in preterm infants: a randomised controlled crossover trial. In this relatively small study of 26 preterm infants who were all less than 32 weeks at delivery, babies following extubation or less invasive surfactant application were randomized to either receive nHFOV then CPAP for four hours each or the reverse order for the same duration. The primary outcome here was reduction in pCO2 with the goal of seeking a difference of 5% or more in favour of nHFOV. Based on their power calculation they thought they would need 24 infants total and therefore exceeded that number in their enrollment.
The babies in both arms were a bit different which may have confounded the results. The group randomized to CPAP first were larger (mean BW 1083 vs 814g), and there was a much greater proportion of males in the CPAP group. As well, the group randomized first to CPAP had higher baseline O2 saturation of 95% compared to 92% in the nHFOV group. Lastly and perhaps most importantly, there was a much higher rate of capillary blood sampling instead of arterial in the CPAP first group (38% vs 15%). In all cases the numbers are small but when looking for such a small difference in pCO2 and the above mentioned factors tipping the scales one way or the other in terms of illness severity and accuracy of measurement it does give one reason to pause when looking at the results.
No difference was found in the mean pCO2 from the two groups. As expected, pCO2 obtained from capillary blood gases nearly met significance for being higher than arterial samples (50 vs 47; p=0.052). A similar rate of babies had to drop out of the study (3 on the nCPAP first and 2 on the nHFOV side).
In the end should we really be surprised by the results? I do believe that in the right baby who is about to fail nCPAP a trial of nHFOV may indeed work. By what means I really don’t understand. Is it the fact that the mean airway pressure is generally set higher than on nCPAP in some studies? Could it be the oscillatory vibration being a kind of noxious stimulus that prevents apneic events through irritation of the infant?
While traditional invasive HFOV does a marvelous job of clearing out CO2 I have to wonder how the presence of secretions and a nasopharynx that the oscillatory wave has to avoid (almost like a magic wave that takes a 90 degree turn and then moves down the airway) allows much of any of the wave to reach the distal alveoli. It would be similar to what we know of inhaled steroids being deposited 90 or so percent in the oral cavity and pharynx. There is just a lot of “stuff” in the way from the nostril to the alveolus.
This leads me to my conclusion that if it is pCO2 you are trying to lower, I wouldn’t expect any miracles with nHFOV. Is it totally useless? I don’t think so but for now as a respiratory modality I think for the time being it will continue to be “looking for a place to happen”
This is becoming “all the rage” as they say. I first heard about the strategy of feeding while on CPAP from colleagues in Calgary. They had created the SINC * (Safe Individualized Feeding Competence) program to provide an approach to safely introducing feeding to those who were still requiring CPAP. As news of this approach spread a great deal of excitement ensued as one can only imagine that in these days when attainment of oral feeding is a common reason for delaying discharge, could getting an early start shorten hospital stay? I could describe what they found with the implementation of this strategy but I couldn’t do it the same justice as the presenter of the data did at a recent conference in Winnipeg. For the slide set you can find them here. As you can imagine, in this experience out of Calgary though they did indeed find that wonderful accomplishment of shorter hospital stays in the SINC group. We have been so impressed with the results and the sensibility of it all that we in fact have embraced the concept and introduced it here in both of our units. The protocol for providing this approach is the following.
I have to admit, while I have only experienced this approach for a short time the results do seem to be impressive. Although anecdotal a parent even commented the other day that she felt that SINC was instrumental in getting her baby’s feeding going! With all this excitement around this technique I was thrown a little off kilter when a paper came out suggesting we should put a full stop to feeding on CPAP!
What caused my spirits to dampen? This study enrolled preterm infants who were still on CPAP at ≥ 34 weeks PMA and were taking over 50% of required feeding volumes by NG feeding. The goal was to look at 15 patients who were being fed on CPAP +5 and with a mean FiO2 of 25% (21-37%) using video fluoroscopic swallowing studies to determine whether such patients aspirate when being fed. The researchers became concerned when each of the first seven patients demonstrated abnormalities of swallowing function indicating varying degrees of aspiration. As such they took each patient off CPAP in the radiology suite and replaced it with 1 l/min NP to achieve acceptable oxygen saturations and repeated the study again. The results of the two swallow studies showed remarkable differences in risk to the patient and as such the recruitment of further patients was stopped due to concerns of safety and a firm recommendation of avoiding feeding while on CPAP was made.
Table 2. Percentage of all swallows identified with swallowing dysfunction
Variable Mean ± s.d.
Mean ± s.d.
Mean ± s.d.
Mild pen. %
Deep pen. %
Nasopharyngeal reflux %
Taking these results at face value it would seem that we should put an abrupt halt to feeding while on CPAP but as the saying goes the devil is in the details…
CPAP Using Ram Cannulae
Let me start off by saying that I don’t have any particular fight to pick with the RAM cannulae. They serve a purpose and that is they allow CPAP to be delivered with a very simple set of prongs and avoid the hats, straps and such of more traditional CPAP devices. We have used them as temporary CPAP delivery when moving a patient from one area to another. As the authors state the prongs are sized in order to ensure the presence of a leak. This has to do with the need to provide a way for the patient to exhale when nasal breathing. Prongs that are too loose have a large leak and may not deliver adequate pressure while those that are too tight may inadvertently deliver high pressure and therefore impose significant work of breathing on the patient. Even with appropriate sizing these prongs do not allow one to exhale against a low pressure or flow as is seen with the “fluidic flip” employed with the infant flow interface. With the fluidic flip, exhalation occurs against very little resistance thereby reducing work of breathing which is not present with the use of the RAM cannula.
Trying to feed an infant who is working against a constant flow as delivered by the RAM cannulae is bound to cause problems. I don’t think it should be a surprise to find that trying to feed while struggling to breathe increases the risk of aspiration. Similarly, under treating a patient by placing them on nasal prongs would lead to increased work of breathing as while you may provide the needed O2 it is at lower lung volumes. Increasing work of breathing places infants at increased risk of aspiration. That is what I would take from this study. Interestingly, looking at the slide set from Calgary they did in fact use CPAP with the fluidic flip. Smart people they are. It would be too easy to embrace the results of this study and turn your nose to the SINC approach to feeding on CPAP. Perhaps somewhere out there someone will read this and think twice about abandoning the SINC approach and a baby will be better for it.
* SINC algorithm and picture of the fluidic flip courtesy of Stacey Dalgleish and the continued work of Alberta Health Services
I had a chance recently to drive a Tesla Model S with autopilot. Taking the car out on a fairly deserted road near my home I flicked the lever twice to activate the autopilot feature and put my hands behind my head while the vehicle took me where I wanted to go. As I cruised down the road with the wheel automatically turning with the curves in the road and the car speeding up or slowing down based on traffic and speed limit notices I couldn’t help but think of how such technology could be applied to medicine. How far away could the self driving ventilator or CPAP device be from development?
I have written about automatic saturation adjustments in a previous post but this referred to those patients on mechanical ventilation. Automatic adjustments of FiO2. Ready for prime time? Why is this goal so important to attain? The reasoning lies in the current design trends in modern NICUs. We are in the middle of a large movement towards single patient room NICUs which have many benefits such as privacy which may lead to enhanced breastfeeding rates and increased parental visitation. The downside, having spoken to people in centres where such designs are already in place is the challenge nursing faces when given multiple assignments of babies on O2. If you have to go from room to room and a baby is known to be labile in their O2 saturations it is human nature to turn the O2 up a little more than you otherwise would to give yourself a “cushion” while you are out of the room. I really don’t fault people in this circumstance but it does pose the question as to whether in a few years we will see a rise in oxygen related tissue injury such as CLD or ROP from such practice. In the previous post I wrote about babies who are ventilated but these infants will often be one to one nursed so the tendency to overshoot the O2 requirements may be less than the baby on non- invasive ventilation.
A System For Controlling O2 Automatically For Infants on Non-Invasive Ventilation
The study was really a proof of concept with 20 preterm infants (mean GA 27.5 weeks, 8 days of age on average) included who each underwent two hours of manual control by nursing to keep saturations between 90-94% and then 4 hours of automated control (sats 91 – 95%) then back to manual for two hours. The slightly shifted ranges were required due to the way in which midpoint saturations are calculated. The essential setup was a computer equipped with an algorithm to make adjustments in FiO2 using an output to a motor that would adjust the O2 blender and then feedback from an O2 saturation monitor back to the computer. The system was equipped with an override to allow nursing to adjust in the event of poor signal or lack of response to the automatic adjustment.
The results though demonstrate that the system works and moreover does a very good job! The average percentage of time that the saturations were in the target range were significantly better with automated control (81% automated, 56% manual). As well as depicted in the following figure the amount of time spent in both hypoxic and hyperoxic ranges was considerable with manual control but non-existent on either tail with automated control (defined as < 85% or > 98% where black bars are manual control and white automatic).
From the figure you can see that the amount of time the patients are in target range are much higher with automatic control but is this simply because in addition to automatic control, nurses are “grabbing the wheel” and augmenting the system here? Not at all.
“During manual control epochs, FiO2 adjustments of at least 1% were made 2.3 (1.3–3.4) times/hour by bedside staff. During automated control, the minimum alteration to FiO2 of 0.5% was being actuated by the servomotor frequently (9.9 alterations/min overall), and changes to measured FiO2 of at least 1% occurred at a frequency of 64 (49–98) /hour. When in automated control, a total of 18 manual adjustments were made in all 20 recordings (0.24 adjustments/hour), a reduction by 90% from the rate of manual adjustments observed during manual control (2.3/hour).”
From the above quote from the paper it is clear that automated control works to keep the saturation goal through roughly 7 X the number of adjustments than nursing makes per hour. It is hard to keep up with that pace when you have multiple assignments but that is what you need I suppose! The use of the auto setting here reduced the amount of nursing interventions to adjust FiO2 by 90% and yields tighter control of O2 saturations.
Dare to Dream
Self driving oxygen administration is coming and this proof of concept needs to be developed and soon into a commercial solution. The risk of O2 damage to developing tissues is too great not to bring this technology forward to the masses. As we prepare to move into a new institution I sincerely hope that this solution arrives in time but regardless I know our nurses and RRTs will do their best as they always do until such a device comes along. When it does imagine all of the time that could be devoted to other areas of care once you were able to move away from the non-invasive device!