Background and aims Although sedation for endotracheal intubation of infants is widely adopted, there is no consensus whether sedation should be used for minimal invasive surfactant therapy (MIST). We compared, in a randomised controlled setting, the level of stress and comfort of preterm infants during MIST with and without receiving low-dose sedation.
Methods Infants between 26 and 36 weeks gestational age were randomised to receive either low-dose sedation (1 mg/kg propofol intravenous) or no premedication during MIST procedure. Standard comfort care was given in both groups, which consisted of administering sucrose in the cheek pouch of the infant and containment. Primary end point was the percentage of infants assessed to be comfortable during the procedure (COMFORTneo-score <14). Secondary parameters included complications of both the MIST procedure and low-dose sedation administration.
Results In total, 78 infants were randomised and analysed, with a median (IQR) gestational age of 29+0 (28+0–32+0) weeks. The percentage of infants with a COMFORTneo score <14 during MIST was significantly higher in the sedated group (32/42 (76%) vs 8/36 (22%), p<0.001). The incidence of desaturation (SpO2 <85%) during the procedure was significantly higher in the sedated group (38/42 (91%) vs 25/36 (69%), p=0.023), and infants needed more often nasal intermittent mandatory ventilation during the procedure (39/42 (93%) vs 17/36 (47%), p<0.001). There were no differences in incidence of hypotension, bradycardia, intubation or pneumothoraxes.
Conclusion Low-dose sedation increased comfort during MIST procedure in preterm infants, but the need for transient non-invasive ventilation was increased.
Trial registration number NTR5010, pre-results.
- preterm infant
- minimal invasive
Statistics from Altmetric.com
What is already known on this topic?
Mechanical ventilation can be avoided by minimally invasive surfactant therapy (MIST), where surfactant is administered to spontaneously breathing infants who then remain on continuous positive airway pressure.
Use of a laryngoscope can lead to cardiovascular responses, and attempts to resist the laryngoscope might result in an increased risk of intraventricular haemorrhage.
There is still no consensus on whether or not to use sedation during MIST.
What this study adds?
Preterm infants receiving low-dose sedation were more often comfortable during the MIST procedure while more often desaturation occurred and nasal intermittent mandatory ventilation was given.
Low-dose sedation could help increase comfort during MIST and might reduce the risk for complications due to stress of a painful stimulus.
Many preterm infants have respiratory distress syndrome due to surfactant deficiency.1 Although nasal continuous positive airway pressure (CPAP) is effective as the initial means of respiratory support in most premature infants, a proportion of infants require surfactant therapy in order to succeed on CPAP.2 3 However, this has traditionally involved intubation followed by mechanical ventilation. Avoiding mechanical ventilation has the potential to decrease the risk for intraventricular haemorrhage (IVH), bronchopulmonary dysplasia and death.4 5 This can be achieved with minimally invasive surfactant therapy (MIST), where surfactant is administered to a spontaneously breathing infant who then remains on CPAP.6 7 The most common method for this approach involves visualising the vocal cords with a laryngoscope and catheterising the trachea using a semirigid catheter, after which the surfactant is instilled.8 Currently, many neonatal intensive care unit (NICU) centres, including in the Netherlands, have adopted this procedure.
There is general consensus that an infant should be adequately sedated when an endotracheal intubation procedure is performed.9–12 Although it is known that visualising the vocal cords with a laryngoscope is highly uncomfortable,13 there is still no consensus on whether or not to use sedation during MIST, particularly when a laryngoscope is used. The use of a laryngoscope can lead to cardiovascular responses, and attempts to resist the laryngoscope might result in an increased risk of intraventricular haemorrhage by impairing cerebral venous return.14–17 While sedation might also increase the chance for an uneventful, smooth and successful procedure, sedation might compromise the infant’s respiratory drive which is a prerequisite for MIST. Reported experience from feasibility studies suggests that the MIST procedure is generally well tolerated without any premedication.1 6 8 However, Klotz et al reported in a European survey that almost half of the neonatologists use premedication during MIST procedures, including sedatives.18
We have performed a randomised controlled trial where we hypothesised that low-dose sedation would increase the comfort of preterm infants during the MIST procedure with minimal side effects. The aim of this study was to compare the level of comfort of preterm infants receiving low-dose sedation versus no sedation during MIST procedure.
A randomised controlled trial was conducted at the Leiden University Medical Center (LUMC), including all preterm infants with a gestational age (GA) between 26 and 37 weeks needing surfactant therapy for respiratory distress syndrome according to the local criteria (FiO2 >0.3 and Positive End Expiratory Pressure ≥8 cm H2O). Opaque sealed envelopes were used to determine randomisation: low-dose sedation or no premedication. We excluded infants who had an imminent need for intubation because of respiratory insufficiency, expressed by apnoea and/or persistent acidosis. Infants with a pneumothorax or pulmonary haemorrhage were also excluded. All infants <30 weeks GA received caffeine in the delivery room or at admission to the NICU. Infants >30 weeks GA received caffeine in case of recurrent apnoea. Allocation was stratified by GA (26–31+6 and 32–36+6 weeks) using variable block sizes (4–8).
The intervention consisted of the use of administering intravenous propofol (1 mg/kg) for sedation during the MIST procedure. This is a reduced dose compared with the standard dose of 2.5 mg/kg used for neonatal intubation to prevent side effects as respiratory depression. Propofol was administered intravenously either by peripheral or central vein dependent on the intravenous access point of the infant. The administration was performed slowly directly before MIST to refrain from thoracic rigidity and pain at administration point. Both groups received standard comfort care, which consisted of administering sucrose 24% in the cheek pouch of the infant and containing the infant during the MIST by swaddling the infant or gently placing the hands of a caregiver on the infant’s body.
The main study parameter was the COMFORTneo score; the primary end point was the percentage of infants with a COMFORTneo score <14 during the procedure. The COMFORTneo score is a validated instrument for measuring objectively the (dis)comfort of a preterm infant19 and is used by trained NICU nurses as standard of care to assess the comfort of a preterm infant every shift at the NICU of the LUMC. The procedure was video recorded, where both the face of the infant and the motions of one of the extremities could be observed, while the other extremities were contained. The recordings were coded and edited so that the administration of sedation was not visible, and two independent NICU nurses, blinded for the allocation, reviewed the recordings and measured the level of comfort using the COMFORTneo scale. Both nurses assessed comfort in 10 MIST procedures, after which inter-rater reliability between these nurses was assessed. Cohen’s κ was used to assess inter-rater reliability, where 0 means no agreement and 1 means total agreement. A Cohen’s κ >0.4 was considered reasonable. When inter-rater reliability appeared to be <0.4, an additional NICU nurse assessed the same 10 MIST procedures to find reasonable reliability between two NICU nurses. When reasonable Cohen’s κ was achieved within two NICU nurses, the remaining MIST procedures were subdivided into those two nurses for assessment.
Secondary study parameters were the occurrence of positive pressure ventilation (nasal intermittent mandatory ventilation (nIMV)) during and immediately after the procedure, intubation need during the procedure and within 24 hours, number of attempts of insertion of the angio-catheter, duration of the procedure, complications occurring during the procedure (desaturation <85%, hypotension with mean below GA, bradycardia <80 bpm, nasal haemorrhage), other complications (pneumothorax, pulmonary haemorrhage, resuscitation, intraventricular haemorrhage ≥grade 3, death), and heart rate and blood pressure before during and after the procedure. Oxygenation and heart rate were measured with the Massimo SET pulse oximeter. Arterial blood pressure was measured using a IntelliVue MP30 Philips Monitor. If an arterial line was not present, blood pressure was measured non-invasively using an appropriate sized neonatal cuff (Philips). All clinical parameters were stored every minute in the local Patient Data Management System (Metavision, IMDSoft, Tel Aviv, Israel).
As there were no data available on comfort during MIST to base the sample size on, we based our sample size calculation on a study that compared intubation during sevoflurane anaesthesia with awake intubation in preterm infants.20 In this trial, the incidence of hypertension, which is a sign of discomfort, was 25% in the anaesthetised infants compared with 56% in awake infants. To detect a comparable decrease in the incidence of COMFORTneo score >14 when using propofol, with a power of 80% and an α error of 5% (two-tailed test), we required 39 infants for each arm.
However, when MIST was imminent and there was no time to approach parents for consent or this was considered inappropriate, deferred consent was asked at a more appropriate time after the procedure. This study was registered in www.trialregister.nl, with registration number NTR5010.
Statistical analysis was performed with SPSS software V.23.0. Demographics of the sedation and the non-sedated group were compared by χ2 test, Student’s t-test and Mann-Whitney U test based on normality of the data. Linear study parameters that were assessed once per infant were compared by a two-way factorial analysis of variance or a linear mixed-effect regression model in multiple assessments, including both the randomisation and stratification group. Categorical outcomes were assessed by Fisher’s exact test. Two-sided p values<0.05 were considered statistically significant.
A total of 110 eligible infants were admitted to the NICU between January 2015 and July 2017, of which 22 infants were not randomised (figure 1). However, four infants in the sedated group and six infants in the non-sedated group were excluded from analysis. A total of 78 infants were analysed, of which 42 in the sedated group and 36 in the non-sedated group. There were no differences between the groups in gestational age, birth weight or gender (table 1).
Time point after birth when MIST procedure was performed was not different (sedated vs non-sedated group: median (IQR) of 5.5 (3–15) vs 6.5 (3–14) hours; p=0.7). The number of attempts needed to insert the angio-catheter in the trachea was not different (1 (1–2) vs 1 (1 – 2) attempts; p=0.982). There was no difference in total duration of the MIST procedure (246.1±174.8 vs 246.1±178.4 s; p=0.641).
The percentage of infants with a COMFORTneo score <14 during MIST was significantly higher in the sedated group compared with the non-sedated group (32/42 (76%) vs 8/36 (22%), p<0.001).
The mean ±SD COMFORTneo score during MIST was significantly lower in the sedated group compared with the non-sedated group (12±3 vs 17±4; p<0.001).
The incidence of desaturation was significantly higher in the sedated group (38/42 (91%) vs 25/36 (69%), p=0.023). There was no difference in mean blood pressure in the 30 min before MIST, during MIST and 30 min after MIST (table 2). The incidence of hypotension was not significantly different between the groups (p=0.282), as was volume expansion as treatment of hypotension (p=1.000) (table 2). In both groups, heart rate was significantly lower during and after the procedure compared with before the MIST procedure. In the sedated group, the difference in heart rate was significantly greater between the periods before, during and after MIST compared with the non-sedated group (p=0.002). There was no difference in the occurrence of bradycardia (heart rate <100 bpm) (p=0.556) (table 2).
More infants in the sedated group needed nIMV during MIST (39/42 (93%) vs 17/36 (47%), p<0.001), but there was no difference in duration of nIMV given (median (IQR) time 7 (3 – 21) vs 6 (3–12) min, p=0.274). Applying nIMV was not influenced by the different GA strata (figure 2).
Only one infant in the sedated group was intubated during MIST, while four infants were intubated in the non-sedated group (1/42 (2%) vs 4/36 (11%), p=0.175). Overall, the rate of intubation in the first 24 hours after MIST was similar in the two groups (10/42 (24%) vs 6/36 (17%), p=0.576). GA stratum did not influence intubation rate.
The incidence of pneumothorax (3/42 (7%) vs 1/36 (3%), p=0.620) and pulmonary haemorrhage (1/42 (2%) vs 0/36 (0%), p=1.000) was not different. There were no differences in rates of intraventricular haemorrhage ≥grade 3 or mortality. None of the included infants needed resuscitation during the MIST procedure.
In this randomised study, we observed that during the MIST procedure more infants were comfortable and the average COMFORTneo score was lower when low-dose sedation was given. However, infants that received sedation were more likely to desaturate and receive nIMV. There were no differences in other complications of the procedure or the use of sedation. This implies that when accepting a temporary need for nIMV low-dose sedation could be used to increase the comfort of preterm infants during the MIST procedure without major clinical implications.
Resistance of the infant during laryngoscopy can lead to an increased risk of intraventricular haemorrhage by impairing cerebral venous return.14–17 Therefore, more comfort during the procedure might be beneficial. The results of the meta-analyses of studies where no sedation is used are conflicting.21 22 While Kribs et al and Klebermass-Schrehof et al 22 described a decrease in IVH, Aldana-Aguirre et al reported no reduction in IVH. We have reported no differences in rate of IVH between the randomisation groups, yet this trial was not powered on this outcome.
Cruz et al reported that infants in the NICU experience many invasive procedures each day, with the highest frequency in the vulnerable first week of life.23 Thereby, infants with the lowest GA and at the highest risk for neurological impairment received the lowest amount of analgesic interventions.23 Procedural pain can affect neurodevelopment as the exposure to multiple painful procedures can lead to ongoing stress.24 There is evidence that these adverse events can be prevented or minimised by using both analgesic and non-pharmacological interventions during painful procedures.25 26 On the other hand, the use of analgesia can lead to adverse effects in the brain as well.27 A high level of analgesia correlates strongly with reduced cerebellar volume and poorer cognitive and motor outcomes in infancy.28–30 However, in our study we used a single low dose of propofol, which reduced the stress of a painful stimulus in this vulnerable population and thereby might have influenced neurodevelopment rather positively than negatively.
While this is the first trial focusing on sedation during MIST, the use of sedation for administrating surfactant by INSURE (intubation–surfactant–extubation) has previously been studied.31 In both the INSURE and the MIST procedures, the sedative agent should have a rapid distribution and redistribution.31 32 Remifentanil is described to have these characteristics and is therefore used in many clinical applications in neonates.33 However, remifentanil was found to be unsuitable as premedication for INSURE as it did not provide adequate sedation.32 In addition, remifentanil is a potent respiratory depressant, which makes it unsuitable as a sedative for MIST.33
Propofol, however, has been widely used as well during intubation and other procedures for sedation and anaesthesia.34–36 Thereby, the level of sedation achieved by the use of propofol was considered satisfactory during surgery in the study of Piersigilli et al, but this was not evaluated objectively.35
One of the most noted side effects of propofol is the occurrence of hypotension, although the incidence varies in studies. Hypotension was not reported in the study which used a dose of 2 mg/kg propofol intravenously,37 while hypotension was reported in studies with an even lower dose of propofol.38 39 In our study, the observed rate of hypotension is similar to our previous observational report.40 In most infants, hypotension was mild and transient and only 3/9 infants (33%) needed a fluid bolus as treatment.
Only 44% of infants receiving MIST in the study of Dargaville et al needed PPV, which we confirm in our non-sedated group (47% needed nIMV). However, we report a significantly higher incidence of need for nIMV in the sedation group (93%). This is due to a higher incidence of desaturation, which was comparable to the study of Descamps et al.41 However, the incidence of need for intubation during or within 24 hours after the procedure did not differ between the randomisation groups, thereby indicating that the need for nIMV was transient.
Intermittent hypoxaemia is associated with an impaired neurodevelopmental outcome.42 However, infants born preterm are intrinsically at risk for intermittent hypoxic episodes during the first 6–8 weeks after birth due to immaturity and many other factors, with a mean of 100–800 hypoxic episodes occurring each week.42 It is therefore difficult or even impossible to differentiate the effect of the single hypoxic episode during MIST procedure on outcome.
The neonatologists performing MIST in this trial were not blinded for the allocated treatment. Although there was clinical equipoise, awareness of the treatment allocation could have influenced the use of nIMV. In contrast, the NICU nurses who analysed the primary outcome were blinded; therefore, this parameter was not influenced by the study design.
Another limitation of this study is the insufficient sample size to perform subanalyses based on different GA strata. However, as we have included GA strata in the models for statistical analyses, our results are corrected for GA.
In this randomised controlled trial, we observed that preterm infants were more often comfortable during the MIST procedure while more often desaturation occurred and nIMV was given. The low-dose sedation could help increase comfort during MIST and might reduce the risk for neurodevelopmental complications due to stress of a painful stimulus. However, only a large and adequately powered randomised trial on this significant clinical outcome can be conclusive on this statement.
The authors thank Patricia Vlasman, Gerben Stigter and Madeleine Olivier for their invaluable assessments of COMFORTneo scores of included infants.
Contributors JD was the executive researcher of this study. She performed literature search, data collection, data analysis, data interpretation, writing and submitting of the report. EL was involved in funding, study design, data interpretation and critical revision of the report. HAvZ was involved in data collection and critical revision of the report. RNGBT was involved in data interpretation and critical revision of the report. SBH was involved in data interpretation and critical revision of the report. AbtP was the project leader and performed literature search, designed the study, and coordinated data analysis, data interpretation, writing and editing of the report. All authors gave approval for the final version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of a part of the work are appropriately investigated and resolved.
Funding Both Chiesi Pharmaceuticals and the Leiden University Fund (LUF) provided an unrestricted grant for the project.
Disclaimer Both Chiesi Pharmaceuticals and the LUF had no role in study design; collection, analysis and interpretation of data; writing of the report; and decision to submit the paper for publication.
Competing interests None declared.
Patient consent Obtained.
Ethics approval The ethical committee of the LUMC approved the study protocol.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.