Methods of surfactant administration and early ventilation in neonatal intensive care units in New South Wales and the Australian Capital Territory

Abstract

BACKGROUND:

This study investigates trends in methods of surfactant administration and early respiratory management in neonatal intensive care units [NICU] in New South Wales [NSW] and the Australian Capital Territory [ACT] in 2015 and evaluate whether differences in practice translate to variances in short term outcomes.

METHODS:

Surveys were sent to NICUs in NSW and ACT to ascertain their practice of surfactant administration and respiratory management. A retrospective data analysis with data from the NICUS database from 01/01/2013-30/06/2015 was performed. Included were all patients that received Surfactant, were inborn, without major malformation, ≥24 weeks gestational age [GA] and birthweight ≥500 g. Major respiratory outcome measures were time ventilated, air leak, oxygen requirement at 36 weeks corrected gestational age [cGA], home oxygen therapy after discharge and retinopathy of prematurity [ROP]. Along with this data demographic and morbidity data was also obtained for comparison [mortality, necrotizing enterocolitis [NEC], persistent ductus arteriosus [PDA], intraventricular hemorrhage [IVH].

RESULTS:

1453 patients met inclusion criteria. Patient data comparing major respiratory outcomes showed patients receiving less invasive Surfactant therapy and respiratory management spent longer time on CPAP [559 vs. 407 hrs, p = 0.01] and in the older gestation subgroups less time on mechanical ventilation [18 vs. 50 hrs p = <0.001] and were discharged earlier [48 vs. 54 days, p = 0.03]. There was however, higher rates of oxygen requirement at 36 weeks cGA [33 vs. 26.3% p = 0.01] and a higher proportion of home oxygen in this patient group [11.3 vs. 7.1% p = 0.03]. Major morbidity outcome data showed no significant differences.

CONCLUSIONS:

Less invasive Surfactant therapy and gentle early respiratory management should be considered as a viable alternative to established methods of surfactant administration and ventilation.

Keywords

Surfactant administration INSURE MIST preterm ventilation

1 Introduction

The use of Surfactant has changed the outcome, morbidity and mortality of premature babies dramatically over the past decades [1, 2]. In order to administer Surfactant safely and effectively it was long believed to be necessary to intubate and subsequently ventilate the baby [3]. Many studies have shown adverse outcomes for babies requiring intubation and ventilation - as opposed to CPAP alone [4 –6]. As this procedure includes the use of narcotic analgesic agents and neuromuscular blocking agents it carries risks of short and long term side-effects; the loss of recruitment of lung volume following neuromuscular blockade along with the risk of neurodevelopmental deficit [7, 8].

Many experimental ways of administering Surfactant without intubation have been investigated subsequently [9, 10]. Since the first reports on minimally invasive Surfactant administration in 2012 many more studies have shown its feasibility [11] and improved outcome in premature infants, where intubation and mechanical ventilation could be avoided [12, 13].

A study published in 2013 by Dargaville et al., from Tasmania, Australia showed a promising alternative way of safely administering Surfactant while the patient was spontaneously breathing on CPAP. This method has been termed Minimally Invasive Surfactant Therapy [MIST] and appears to have the potential to alter respiratory course and outcome [14]. At the same time large multi-centre trials mostly in Nordic Europe and Germany have been carried out showing further benefits of this less invasive technique [12 , 16].

Beltempo et al., published international survey results of the respiratory management of preterm infants worldwide showing a wide variety in approach to ventilation strategies used in preterm infants [17]. Detailed and current data on the practice of Surfactant administration and approach to ventilation in neonates and whether the different approach leads to a difference in outcome in Australian NICUs are outstanding to date.

2 Materials and methods

2.1 Survey and database

To establish the current protocols for Surfactant administration in NSW and ACT surveys were sent out to the head of department at the 8 perinatal NICUs in these states. The 8 participating Intensive Care units are sited at the Canberra Hospital, John Hunter Children’s Hospital, Liverpool Hospital, Nepean Hospital, Royal Prince Alfred Hospital, Royal Hospital for Women, Royal North Shore Hospital, and Westmead Hospital.

The survey answers were then compared with the recorded practise using retrospective data from the Neonatal Intensive Care Units’ Data Collection [NICUS]. This database is an ongoing prospectively collected regional audit tool for live born neonates admitted to NICUs in NSW and ACT. Included in the database are neonates, who meet any of the following eligibility criteria: gestation < 32 weeks and/or birth weight < / = 1500 g and/or need for assisted ventilation [mechanical ventilation, continuous positive airways pressure [CPAP], high flow oxygen/air > 1 L/min] and/or major surgery [opening of a body cavity] and/or insertion of a central line and/or intentional hypothermia for infants with hypoxic ischemic encephalopathy, and/or exchange transfusion [18].

To determine safety, efficacy and differences in outcome of minimally invasive Surfactant administration and non-invasive ventilation strategies this study retrospectively enrolled patients admitted between 01/01/2013 and 30/06/2015.

2.2 Patients

This study investigated the outcome of all babies admitted to the NICU that received Surfactant therapy, were inborn [born in a perinatal NICU in NSW or ACT], without major malformation, 24 weeks gestation and above and 500 g birth weight and above. Out of 8084 admitted patients 1453 patients met the inclusion criteria. Figure 1 shows a flowchart describing the inclusion process.

Fig.1

Flowchart of excluded cases.

The patients were divided into groups depending on whether they were treated in a unit with less invasive respiratory management or more invasive respiratory management. The units were divided into two groups according to a combination of survey results and actual practice as evidenced by an audit. There were four units whose stated approach and audit results matched – two who preferred intubation/ventilation and on audit clearly managed their babies by that method, and two who having stated a preference for MIST/INSURE demonstrated this approach in actual practice by using MIST/INSURE in more than 70% of cases. The other 4 units had a stated preference for INSURE but on audit, were shown to use that approach in less than 40% of cases. Therefore, for the purposes of this paper, only the two units for whom stated preference matched audit results are regarded as less invasive units and the four units with variance between stated and measured practice are regarded as more invasive units, so were the two units where survey and audit matched.

The two NICUs deemed less invasive included 330 patients of which 170 patients were treated in one NICU facilitating MIST as Surfactant administration technique and 160 in another NICU facilitating INSURE. The remaining 6 units treated the remaining 1123 patients.

A comparison of patients divided by Surfactant administration technique alone [MIST vs. INSURE, vs. Prolonged mechanical ventilation] irrespective of the NICU where they were managed revealed significant differences in demographics and condition at birth. An analysis of differences in outcome was therefore, futile and the above mentioned method of grouping remained as the only viable analysis.

To determine whether gestational age was a major contributor to differences in outcome depending on respiratory management and Surfactant administration, the patients were further subdivided into three groups [24–28 weeks GA vs. 28–32 weeks GA vs > 32 weeks GA]. This was especially useful as the audit revealed more invasive units’ practice of preselection of infants born at a later gestation to preferentially receive less invasive Surfactant administration methods [i.e. INUSRE] and spent a shorter period of time on mechanical ventilation.

Respiratory outcomes investigated were time ventilated [mechanical ventilation and CPAP], air leak [pneumothorax], oxygen requirement at 36 weeks corrected gestational age in infants born before 32 weeks GA, home oxygen therapy after discharge, pulmonary hemorrhage, and retinopathy of prematurity. Along with this data, demographic [gender, antenatal steroids, 1 and 5 minutes Apgar, plurality, birthweight, gestational age, prolonged rupture of membranes] and morbidity data [mortality, NEC, PDA, IVH, inotrope use at 48 hours of life] was also obtained to compare the groups. Ethics Approval was sought from the South Western Sydney Local Health District HREC [EC00136] ethics committee for low risk studies [de-identified data].

2.3 Statistical analysis

Univariate and multivariate analysis was used facilitating SPSS version 13.0. Two sided T-tests were used to analyse potential differences between the study groups. P values of < 0.05 were considered statistically significant.

3 Results

3.1 Survey

All enrolled NICUs responded with a complete survey within a timeframe of four weeks. All surveys were returned by 30/06/2015. The surveys confirmed only one NICU in NSW and ACT administers Surfactant via MIST. Five out of eight NICUs reported INSURE to be their preferred method of Surfactant administration. Two units reported back to prefer intubation and ventilation for longer than one hour post Surfactant administration. Most departments described some method of preselection for premature infants for INSURE or ventilation with gestational age and condition at birth being the most common determining factor.

There was a difference in protocol and perception vs documented practise with only four units actually employing their protocol as stated in the survey for the method of Surfactant administration in the majority of patients. Unit code 2 and 4 both answered main method of Surfactant administration via intubation and ventilation for > 1hr and documented number of patients following this approach 78.8% and 78.9% respectively. Unit code 8 treated 79.4% of their patients with INSURE their proclaimed main method of Surfactant administration technique and unit code 5 using MIST in 80% of their patients. Unit code 3 identified INSURE as their main administration technique but in fact 82.1% of patients remained intubated for a prolonged period of time and Units code 1, 6 and 7, having replied the same, ventilated more than 60% of their patients after Surfactant administration [64.2%, 61.6%, 67.5% ].

Interestingly, when analysing the data for premature infants born before 27 weeks six of the eight NICUs elect to intubate and ventilate rather than using less invasive techniques for Surfactant administration [between 97.1% and 84% of the patients], indicating preselection of patients for less invasive Surfactant administration and ventilation strategy contributing to aforementioned significant differences in patient population. “Table 1” shows demographic data of patients divided by Surfactant administration technique alone. The audit of the remaining two NICUs resulted in comparably low rates of mechanical ventilation and comparably high rates of patients receiving less invasive Surfactant administration in both patients born before and after 27 weeks [67.3% and 81.8% in patients < 27 weeks].

Table 1
Demographics of MIST versus INSURE versus ventilated patients

24–41 weeks GA MIST INSURE p = Ventilated p = MIST p = INSURE

n = 129 437 886

Birth weight [mean in gram] 1468 1447 0.76 1373 0.17 0.05

GA [median in weeks] 29.6 29.5 0.7 28.9 0.13 <0.001ª

GA<27 wks [% ] 25.6 10.5 <0.001ª 39.6 0.06 <0.001ª

Male [% ] 55.8 54.2 0.82 54.6 0.32 0.31

Multiple [% ] 24 29.3 0.18 24.7 0.32 0.28

Apgar 1 minute [mean] 5.8 6.3 0.14 5.1 <0.001ª <0.001ª

Apgar 5 minutes [mean] 7.7 8 0.04ª 7.0 <0.001ª <0.001ª

PPROM > 24 hrs [% ] 26.4 17.4 <0.001ª 26.5 0.92 <0.001ª

Antenatal steroids complete [% ] 59.5 60.5 0.85 60.3 0.13 0.54

24–41 weeks GA	MIST	INSURE	p =	Ventilated	p = MIST	p = INSURE
Birth weight [mean in gram]	1468	1447	0.76	1373	0.17	0.05
GA [median in weeks]	29.6	29.5	0.7	28.9	0.13	<0.001ª
GA<27 wks [% ]	25.6	10.5	<0.001ª	39.6	0.06	<0.001ª
Male [% ]	55.8	54.2	0.82	54.6	0.32	0.31
Multiple [% ]	24	29.3	0.18	24.7	0.32	0.28
Apgar 1 minute [mean]	5.8	6.3	0.14	5.1	<0.001ª	<0.001ª
Apgar 5 minutes [mean]	7.7	8	0.04ª	7.0	<0.001ª	<0.001ª
PPROM > 24 hrs [% ]	26.4	17.4	<0.001ª	26.5	0.92	<0.001ª
Antenatal steroids complete [% ]	59.5	60.5	0.85	60.3	0.13	0.54

ªp = <0.05.

3.2 Patient demographics

The demographic data collected for the different groups showed only one parameters that was statistically significantly different. There were significantly more multiple births in the units with more invasive respiratory management in the group born between 28 and 32 weeks gestation. All other demographic parameters [birthweight, gestational age, gender and antenatal steroids, PPROM] were not significantly different neither in the total group of patients nor in the subgroups. “Table 2” shows the results in detail.

Table 2
Demographics of patients in different age groups divided depending on treatment in NICU with invasive or less invasive respiratory management

24–41 weeks GA Invasive Less invasive p =

n = 1123 330

Birthweight [mean in gram] 1397 1384 0.77

GA [median in weeks] 29.2 29 0.22

Male [% ] 57 54 0.32

Multiple [% ] 25.4 26.7 0.25

Apgar 1 minute [median] 6 5 0.61

Apgar 5 minutes [median] 8 8 0.5

PPROM > 24 hrs [% ] 24.5 23.5 0.7

Antenatal steroids complete [% ] 62.8 61.4 0.99

INSURE/MIST received [% ] 26.6 79.4 <0.001ª

24–28 weeks GA Invasive Less invasive p =

n = 537 167

Birthweight [mean in gram] 940 940 1

GA [median in weeks] 26.5 26.4 0.42

Male [% ] 52.5 52.7 0.98

Multiple [% ] 27.7 30.5 0.51

Apgar 1 minute [median] 5 5 0.05

Apgar 5 minutes [median] 7 7 0.6

PPROM > 24 hrs [% ] 28.5 28.7 0.93

Antenatal steroids complete [% ] 58.2 62.4 0.39

INSURE/MIST received [% ] 17.5 77.8 <0.001ª

28–32 weeks GA Invasive Less invasive p =

n = 431 115

Birthweight [mean in gram] 1504 1519 0.69

GA [median in weeks] 30.4 30.2 0.64

Male [% ] 55.5 64.4 0.07

Multiple [% ] 29 19 0.03ª

Apgar 1 minute [median] 6 6 0.69

Apgar 5 minutes [median] 8 8 0.42

PPROM > 24 hrs [% ] 23 24.3 0.78

Antenatal steroids complete [% ] 65.1 62 0.64

INSURE/MIST received [% ] 38.3 80.9 <0.001ª

>32 weeks GA Invasive Less invasive p =

n = 151 46

Birthweight [mean in gram] 2708 2630 0.55

GA [median in weeks] 35.6 35 0.1

Male [% ] 53.6 52.2 0.92

Multiple [% ] 15 6.5 0.15

Apgar 1 minute [median] 6 8 0.08

Apgar 5 minutes [median] 8 9 0.05

PPROM > 24 hrs [% ] 7.9 10.9 0.59

Antenatal steroids complete [% ] 56 42.3 0.2

INSURE/MIST received [% ] 26.5 95.7 <0.001ª

24–41 weeks GA	Invasive	Less invasive	p =
Birthweight [mean in gram]	1397	1384	0.77
GA [median in weeks]	29.2	29	0.22
Male [% ]	57	54	0.32
Multiple [% ]	25.4	26.7	0.25
Apgar 1 minute [median]	6	5	0.61
Apgar 5 minutes [median]	8	8	0.5
PPROM > 24 hrs [% ]	24.5	23.5	0.7
Antenatal steroids complete [% ]	62.8	61.4	0.99
INSURE/MIST received [% ]	26.6	79.4	<0.001ª
24–28 weeks GA	Invasive	Less invasive	p =
n =	537	167
Birthweight [mean in gram]	940	940	1
GA [median in weeks]	26.5	26.4	0.42
Male [% ]	52.5	52.7	0.98
Multiple [% ]	27.7	30.5	0.51
Apgar 1 minute [median]	5	5	0.05
Apgar 5 minutes [median]	7	7	0.6
PPROM > 24 hrs [% ]	28.5	28.7	0.93
Antenatal steroids complete [% ]	58.2	62.4	0.39
INSURE/MIST received [% ]	17.5	77.8	<0.001ª
28–32 weeks GA	Invasive	Less invasive	p =
n =	431	115
Birthweight [mean in gram]	1504	1519	0.69
GA [median in weeks]	30.4	30.2	0.64
Male [% ]	55.5	64.4	0.07
Multiple [% ]	29	19	0.03ª
Apgar 1 minute [median]	6	6	0.69
Apgar 5 minutes [median]	8	8	0.42
PPROM > 24 hrs [% ]	23	24.3	0.78
Antenatal steroids complete [% ]	65.1	62	0.64
INSURE/MIST received [% ]	38.3	80.9	<0.001ª
>32 weeks GA	Invasive	Less invasive	p =
n =	151	46
Birthweight [mean in gram]	2708	2630	0.55
GA [median in weeks]	35.6	35	0.1
Male [% ]	53.6	52.2	0.92
Multiple [% ]	15	6.5	0.15
Apgar 1 minute [median]	6	8	0.08
Apgar 5 minutes [median]	8	9	0.05
PPROM > 24 hrs [% ]	7.9	10.9	0.59
Antenatal steroids complete [% ]	56	42.3	0.2
INSURE/MIST received [% ]	26.5	95.7	<0.001ª

ªp = <0.05. NICU = Neonatal Intensive Care Unit, MIST = minimally invasive surfactant therapy, GA = gestational age, BW = birth weight, wks = weeks, PPROM = prolonged premature rupture of membranes, INSURE = intubate Surfactant extubate.

3.3 Patient outcomes

Respiratory outcomes of the total population showed significantly longer CPAP times in the two NICUs with less invasive respiratory management [559 hrs vs. 407 hrs]. We also found a significantly higher rate of Oxygen requirement at 36 weeks [33% vs. 26.3% ] and more need for home oxygen in this patient cohort [11.3% vs. 7.1% ]. Time to Surfactant administration was significantly longer in the less invasive NICUs [297 mins vs. 231 mins].

In the gestational age subgroups we found significantly longer times on CPAP for patients born before 28 weeks [903 hrs vs. 681 hrs] and shorter time on mechanical ventilation for babies born between 28 and 32 weeks in the less invasive group [18 hrs vs. 50 hrs]. Time to Surfactant was significantly longer in the < 28 weeker group in less invasive NICUs [119 mins vs. 59.4 mins]. And significantly fewer patients in these NICUs born 28–32 weeks suffered a pulmonary hemorrhage [0% vs. 1.6% ]. However, more patients were discharged home on Oxygen in this group [8.7% vs. 3% ]. Patients treated in the two units with less invasive respiratory management and born after 28 weeks gestation were discharged significantly earlier [48.3 days vs. 53.5 days]. Babies born after 32 weeks had significantly less need for inotropic agents at 48 hrs in the less invasive group [8.7% vs. 21.9% ]. “Table 3” show the results in detail.

Table 3
Outcome of patients in different age groups divided depending on treatment in NICU with

24–41 weeks GA Invasive Less invasive p =

n = 1123 330

O₂ at 36 wks [% ] 26.3 33 0.01ª

Mech vent [hrs] 101.5 83.6 0.2

CPAP [hrs] 407 559 <0.001ª

Home O₂ [% ] 7.1 11.3 0.03ª

Pneumothorax [% ] 5.7 5.5 0.86

Pulm hem [% ] 3.8 3.3 0.66

Time to surfactant [mean in min] 231 297 0.04ª

Discharge age [mean in days] 63.3 61.5 0.45

Inotropes at 48 hrs [% ] 17.3 20.3 0.22

IVH major [% ] 3.1 5.5 0.09

Mortality [% ] 6.6 7.3 0.67

NEC surgery [% ] 1 2.1 0.18

PDA surgery [% ] 2.8 3.6 0.44

ROP requiring laser [% ] 3.6 3.6 0.95

24–28 weeks GA 537 167

n =

O₂ at 36 wks [% ] 44 51.5 0.09

Mech vent [hrs] 151 143 0.76

CPAP [hrs] 681 903 <0.001ª

Home O₂ [% ] 11.9 16 0.18

Pneumothorax [% ] 3.7 3 0.64

Pulm hem [% ] 6.1 6 0.94

Time to surfactant [mean in min] 59.4 119 0.04ª

Discharge age [mean in days] 82.8 83 0.95

Inotropes at 48 hrs [% ] 23 28 0.19

IVH major [% ] 5 6.6 0.52

Mortality [% ] 10.8 11.4 0.83

NEC surgery [% ] 2 4.2 0.2

PDA surgery [% ] 5 7.2 0.32

ROP requiring laser [% ] 7.4 7.2 0.9

28–31 weeks 431 115

n =

O₂ at 36 wks [% ] 13.5 20 0.09

Mech vent [hrs] 50 18 <0.001ª

CPAP [hrs] 189 252 0.07

Home O₂ [% ] 3 8.7 0.04ª

Pneumothorax [% ] 6.3 6.9 0.8

Pulm hem [% ] 1.6 0 0.008ª

Time to surfactant [mean in min] 306 365 0.24

Discharge age [mean in days] 53.5 48.3 0.03ª

Inotropes at 48 hrs [% ] 8.6 13.9 0.13

IVH major [% ] 1.2 5.2 0.06

Died [% ] 1.9 3.5 0.38

NEC surgery [% ] 0 0 n/a

PDA surgery [% ] 0.9 0 0.08

ROP requiring laser [% ] 0 0 n/a

>32 weeks 151 46

n =

Mech vent [hrs] 73 33.5 0.07

CPAP [hrs] 57 93 0.09

Home O2 [% ] 2 0 0.08

Pneumothorax [% ] 11 10.9 0.91

Pulm hem [% ] 2 2.2 0.94

Time to surfactant [mean in min] 622 761 0.26

Discharge age [mean in d] 22.6 16.5 0.02ª

Inotropes at 48 hrs [% ] 21.9 8.7 0.01ª

IVH major [% ] 1.3 2.2 0.72

Mortality [% ] 5.3 2.2 0.26

NEC surgery [% ] 0 0 n/a

PDA surgery [% ] 0.7 0 0.31

ROP requiring laser [% ] 0 0 n/a

24–41 weeks GA	Invasive	Less invasive	p =
O₂ at 36 wks [% ]	26.3	33	0.01ª
Mech vent [hrs]	101.5	83.6	0.2
CPAP [hrs]	407	559	<0.001ª
Home O₂ [% ]	7.1	11.3	0.03ª
Pneumothorax [% ]	5.7	5.5	0.86
Pulm hem [% ]	3.8	3.3	0.66
Time to surfactant [mean in min]	231	297	0.04ª
Discharge age [mean in days]	63.3	61.5	0.45
Inotropes at 48 hrs [% ]	17.3	20.3	0.22
IVH major [% ]	3.1	5.5	0.09
Mortality [% ]	6.6	7.3	0.67
NEC surgery [% ]	1	2.1	0.18
PDA surgery [% ]	2.8	3.6	0.44
ROP requiring laser [% ]	3.6	3.6	0.95
24–28 weeks GA	537	167
n =
O₂ at 36 wks [% ]	44	51.5	0.09
Mech vent [hrs]	151	143	0.76
CPAP [hrs]	681	903	<0.001ª
Home O₂ [% ]	11.9	16	0.18
Pneumothorax [% ]	3.7	3	0.64
Pulm hem [% ]	6.1	6	0.94
Time to surfactant [mean in min]	59.4	119	0.04ª
Discharge age [mean in days]	82.8	83	0.95
Inotropes at 48 hrs [% ]	23	28	0.19
IVH major [% ]	5	6.6	0.52
Mortality [% ]	10.8	11.4	0.83
NEC surgery [% ]	2	4.2	0.2
PDA surgery [% ]	5	7.2	0.32
ROP requiring laser [% ]	7.4	7.2	0.9
28–31 weeks	431	115
n =
O₂ at 36 wks [% ]	13.5	20	0.09
Mech vent [hrs]	50	18	<0.001ª
CPAP [hrs]	189	252	0.07
Home O₂ [% ]	3	8.7	0.04ª
Pneumothorax [% ]	6.3	6.9	0.8
Pulm hem [% ]	1.6	0	0.008ª
Time to surfactant [mean in min]	306	365	0.24
Discharge age [mean in days]	53.5	48.3	0.03ª
Inotropes at 48 hrs [% ]	8.6	13.9	0.13
IVH major [% ]	1.2	5.2	0.06
Died [% ]	1.9	3.5	0.38
NEC surgery [% ]	0	0	n/a
PDA surgery [% ]	0.9	0	0.08
ROP requiring laser [% ]	0	0	n/a
>32 weeks	151	46
n =
Mech vent [hrs]	73	33.5	0.07
CPAP [hrs]	57	93	0.09
Home O2 [% ]	2	0	0.08
Pneumothorax [% ]	11	10.9	0.91
Pulm hem [% ]	2	2.2	0.94
Time to surfactant [mean in min]	622	761	0.26
Discharge age [mean in d]	22.6	16.5	0.02ª
Inotropes at 48 hrs [% ]	21.9	8.7	0.01ª
IVH major [% ]	1.3	2.2	0.72
Mortality [% ]	5.3	2.2	0.26
NEC surgery [% ]	0	0	n/a
PDA surgery [% ]	0.7	0	0.31
ROP requiring laser [% ]	0	0	n/a

ªp = <0.05. O₂= Oxygen, wks = weeks, mech vent = mechanical ventilation, hrs = hours, CPAP = continuous positive airway pressure, pulm hem = Pulmonary hemorrhage, min = minutes, IVH = intraventricular hemorrhage, NEC = necrotizing enterocolitis, PDA = persistent ductus arteriosus, ROP = retinopathy of prematurity, ªp = <0.05.

In an attempt to compare the outcome of patients treated with two different less invasive Surfactant administration techniques the two units with similar approach to respiratory management were compared. Both units aimed for minimal time on mechanical ventilation and for less invasive Surfactant administration techniques. However, one unit facilitated MIST and the other INSURE as their preferred technique. Results from these two NICUs were compared to establish whether there were any major differences in outcome using two different less invasive surfactant administration techniques.

As outlined in “Table 4” the demographics showed no difference apart from one characteristic [more male patients in Unit 8 compared to 5]. The patients’ condition at birth and antenatal risk factor profile was comparable. In terms of clinical outcome the only significantly different results were time to Surfactant [235 vs. 356 minutes] and discharge weight [2899 vs 2621 g] as outlined in “Table 5”

Table 4

Demographics of patients treated in NICUs with less invasive respiratory management facilitating MIST vs INSURE for Surfactant administration

Less invasive NICUs	MIST	INSURE	p =
facilitating n=	170	160
BW [mean g]	1361	1423	0.42
GA [Median in wks]	28	29	0.47
GA<27 wks [% ]	32.4	36.6	0.42
Male [% ]	40.5	58.6	0.012ª
Multiple [% ]	22.9	24	0.76
Apgar 1’ [median]	6	5	0.36
Apgar 5’ [median]	8	8	0.55
Antenatal steroids [% ]	61	58.6	0.319
PPROM > 24 h [% ]	22.9	26	0.48

Table 5

Outcomes of patients treated in NICUs with less invasive respiratory approach facilitating MIST vs INSURE for Surfactant administration

Less invasive NICUs	MIST	INSURE	p =
facilitating n =	170	160
O₂ at 36 wks [% ]	35.8	30	0.30
Mech vent [hrs]	67	101	0.17
CPAP [hrs]	574	555	0.75
Home O2 [% ]	14.1	8.1	0.17
Pneumothorax [% ]	3	8	0.55
Pulm hem [% ]	2.9	3.8	0.55
Time to surfactant [mean in min]	356	235	0.021ª
Discharge weight [mean in g]	2621	2899	0.006ª
Discharge age [mean in d]	61	66	0.2
Inotropes at 48 hrs [% ]	19.4	21.2	0.38
IVH major [% ]	5.9	2.5	0.14
Mortality [% ]	6.5	8.1	0.5
NEC surgery [% ]	1.2	3.1	0.69
PDA surgery [% ]	1.8	5.6	0.44
ROP requiring laser [% ]	2.9	4.4	0.68

4 Discussion

The results of this study show the differences in approach versus policy when it comes to Surfactant administration method. Despite many NICUs aspiring to administer Surfactant in a less invasive way – indicated by their reply on the survey – many do not achieve this aim in reality. A possible reason for the marked difference between perception and reality about Surfactant administration methods could be that the surveys were only sent to one person – the head of the department - and the answers could represent and reflect this person’s perception rather than actual practice.

There are many similarities when comparing the results of the survey used in this study and the ANZNN [Australian and New Zealand Neonatal Network] replies to the international survey conducted by the group around Beltempo in 2018 [17]. The common respiratory management for infants born > 27 weeks GA was CPAP alone with targeted Surfactant therapy for those who qualified. There was also a significant shift reported towards the use of less invasive Surfactant administration techniques like MIST, LISA, and INSURE in this cohort. However, most networks still elect to intubate and ventilate infants born prior to this gestation, again a parallel to the findings in our study.

The results of the group comparison between patients born in units with a less invasive respiratory management approach and patients born in units with more invasive strategies showed some minor differences in outcome, some of which can be explained by the actual treatment itself. Avoiding mechanical ventilation and aiming for CPAP from birth and throughout their stay in hospital will lead to less time on mechanical ventilation whilst spending more time on CPAP as CPAP is used as an alternative to mechanical ventilation. Also less invasive techniques to administer Surfactant will take slightly longer than intubating the patient at birth and administering Surfactant irrespective of their own respiratory effort. Stabilising a premature infant, especially an extremely premature infant on CPAP at birth can require a longer period of time. Lung recruitment manoeuvres are prolonged when only CPAP is offered compared to a pre-emptive intubation at birth and positive pressure ventilation. Another reason for the slightly delayed Surfactant administration could be the lack of premedication which means it is paramount that the patients’ respiratory status is stable as they have to breathe spontaneously on CPAP whilst the procedure is performed. Dargaville et al., found in their feasibility study for MIST that most of their MIST procedures were performed in the NICU rather than in the delivery room leading to a delay in their patients as well [19]. The higher rate of infants born after 28 weeks and being discharged on Oxygen in the less invasive units might be contributed to the earlier discharge date in this patient cohort [on average five days earlier]. The other results were quite variable throughout the different age groups or did not reach statistical significance.

Comparing MIST versus INSURE patients in the NICUs with the most similar initial respiratory management showed no difference in major pulmonary outcomes and morbidity. The results found in this study go in line with multiple studies on this subject making MIST a viable alternative to INSURE [20].

The results of this study are to some extent comparable to research done in this field in the last few years. Gozde Kanmaz et al., performed a randomized study in 2011 enrolling 100 preterm infants < 32 weeks GA for either minimally invasive Surfactant administration or INSURE. They found significantly shorter duration of mechanical ventilation time, similar to the findings in our study. They did however, show lower rates of chronic lung disease [CLD] in the MIST group [13]. Gopel et al., published results from a randomised controlled trial in the Lancet in 2011 having found a reduction in time spent on mechanical ventilation and less time needing supplemental Oxygen in patients receiving less invasive Surfactant administration [16]. The NINSAPP study in 2015 was a randomized controlled trial treating 211 infants < 28 weeks GA with less invasive Surfactant administration. The study showed a significantly higher survival without major complications, particularly a large reduction in major IVH in the less invasive group [12]. And finally a recent meta-analysis of studies involving less invasive Surfactant administration and minimally invasive Surfactant therapy compared to intubation demonstrated a reduction of CLD and death in premature infants [21].

Compared to these studies and analyses our investigation differs. A retrospective data analysis and survey will always be a reflection of actual practice versus practice under research conditions. One of the problems with research conditions is that they are very controlled and not necessarily how these approaches work in every day clinical work. As admitted by some of the investigators of the above mentioned trials study patients might subconsciously be subjected to different clinical care. Mulder et al wrote an opinion piece on exactly this problem in the BMJ in 2017 where he discussed limitations of controlled trials for medication in mental health. According to him the bias in design, recruitment, sample population, and data analysis exacerbate limitations of applying study methods to real-life clinical management [22]. And this paper demonstrates this problem – at least 4 units responded by saying that INSURE was their preferred approach but in everyday practice, it was only utilised in less than 40% of cases.

It is always difficult in a retrospective data analysis to find groups that are truly comparable with the variety of department specific policies on so many different approaches and treatments for premature babies. Every Unit will have a different approach to the treatment of a premature infant and that starts but certainly is not limited to the respiratory management [23 –25]. In our study this was demonstrated in the comparison of patients receiving INSURE versus MIST where significant demographical differences were found which made a meaningful comparison of these two groups impossible. So, instead a comparison of overall approach to ventilation in the first 24 hours was undertaken to overcome this issue.

Therefore, our retrospective study has limitations. However, strict inclusion criteria and a grouping of gestational ages and clinical approaches to initial treatment addressed some of the bias issues and we managed to achieve two groups that were truly comparable.

By comparing two NICUs with a similar approach to minimal intervention and gentle ventilation and Surfactant administration techniques we tried to overcome issues of lack of comparability. Both units perform only limited preselection and reliably use less invasive Surfactant administration [MIST or INSURE] in the majority of their patients.

So in summary, results from NICUs that aim for less invasive respiratory management and Surfactant administration technique for all their premature patients, irrespective of their gestational age or size, are comparable and not inferior to results from patients treated in NICUs with a more invasive approach to Surfactant administration and ventilation.

When reviewing the outcomes of the patients that received MIST, under the premise of no preselection and a truly committed approach to the method, we achieved to show that MIST is an effective method to administer Surfactant and avoid mechanical ventilation. Patients treated with MIST achieved similar results as patients that received INSURE.

The results of this study show MIST is in no way inferior to INSURE, especially in patients born at lower gestational age. This study reinforces the valuable role of less invasive respiratory management in all premature infants.

5 Conclusion

Less invasive Surfactant administration methods and an aim to reduce time spent on mechanical ventilation should be considered as a valuable and safe alternative to established methods of Surfactant administration and respiratory management.

The authors acknowledge the valuable input Dr. Jacqueline Stack made to this manuscript. Significant writing assistance and guidance was sought.

References

Polin

, Carlo

. Surfactant replacement therapy for preterm and term neonates with respiratory distress. Pediatrics. 2014;133(1):156–63.

Milner

. Surfactant and respiratory distress syndrome. Turk J Pediatr. 1996;38(1):37–43.

McPherson

, Wambach

. Prevention and treatment of respiratory distress syndrome in preterm neonates. Neonatal Netw. 2018;37(3):169–77.

Carlo

, Walsh

, Rich

, et al. Early CPAP versus surfactant in extremely preterm infants. N Engl J Med. 2010;362:1970–9.

Flannery

, O’Donnell

, Kornhauser

, Dysart

, Greenspan

, Aghai

Continuous Positive Airway Pressure versus Mechanical Ventilation on the First Day of Life in Very Low-Birth-Weight Infants. Am J Perinatol. 2016;33(10):939–44.

Morley

, Davis

, Doyle

, Brion

, Hascoet

, Carlin

. COIN Trial Investigators Nasal CPAP or intubation at birth for very preterm infants. N Engl J Med. 2008;358(7):700–8.

Bjorklund

, Ingimarsson

, Curstedt

, et al. Manual ventilation with a few large breaths at birth compromises the therapeutic effect of subsequent Surfactant replacement in immature lambs. Pediatr Res. 1997;42(3):348–55.

Flick

, Katusic

, Colligan

, et al. Cognitive and behavioural outcomes after early exposure to anaesthesia and surgery. Pediatrics 2011; 128:e1053–e1061.

Aguar

, Cernada

, Brugada

, Gimeno

, Gutierrez

, Vento

. Minimally invasive Surfactant therapy with a gastric tube is as effective as the intubation, Surfactant, and extubation technique in preterm babies. Acta Paediatr. 2014; 103(6):e229–33.

10.

Abdel-Latif

, Osborn

. Nebulised Surfactant in preterm infants with or at risk of respiratory distress syndrome. Cochrane Database Syst Rev. 201217;10:CD008310.

11.

Mohammadizadeh

, Ghehsareh

, Sadeghnia

. Early administration of Surfactant via a thin intratracheal catheter in infants with respiratory distress syndrome: Feasibility and outcome. J Res Pharm Pract. 2015;4(1):31–6.

12.

Kribs

, Roll

, Gopel

, et al. Nonintubated Surfactant Application vs. Conventional Therapy in Extremely Preterm Infants, A Randomized Clinical trial. JAMA Pediatr. 2015;169(8):723–30.

13.

Gozde Kanmaz

, Erdeve

, Canpolat

, Mutlu

, Dilmen

. Surfactant administration via thin catheter during spontaneous breathing: Randomized controlled trial. Pediatrics 2013; 131;e502.

14.

Dargaville

, Aiyappan

, De Paoli

, et al. Minimally-invasive Surfactant therapy in preterm infants on continuous positive airway pressure. Arch Dis Child Fetal Neonatal Ed. 2013; 98(2):F122–6.

15.

Heiring

, Jonsson

, Andersson

, et al. Survey shows large differences between the Nordic countries in the use of less invasive Surfactant administration. Acta Paediatric. 2016;106:382–6.

16.

Gopel

, Kribs

, Ziegler

, et al. Avoidance of mechanical ventilation by surfactant treatment of spontaneously breathing preterm infants (AMV): An open label, randomised, controlled trial. Lancet. 2011;378:1627–34.

17.

Beltempo

, Isayama

, Vento

, et al. Respiratory Management of Extremely Preterm Infants: An international Survey. Neonatology 2018;114(1):28–36.

18.

[Neonatal Intensive Care Units’ Data Collection, NSW Pregnancy and Newborn Services Network].

19.

Dargaville

, Aiyappan

, Cornelius

, Williams

, De Paoli

. Preliminary evaluation of a new technique of minimally invasive surfactant therapy. Arch Dis Child Fetal Neonatal Ed. 2011; 96(4):F243–8.

20.

Heidarzadeh

, Mirnia

, Hoseini

, et al. Surfactant administration via thin catheter during spontaneous breathing: Randomized controlled trial in Alzahra hospital. Iran J Neonatol. 2013;4:5–9.

21.

Aldana-Aquire

, Pinto

, Featherstone

, Kuma

. Less invasive Surfactant administration versus intubation for Surfactant delivery in preterm infants with respiratory distress syndrome: A systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed 2017; 102:F17–F23.

22.

Mulder

, Singh

, Hamilton

, et al. The limitations of using randomised controlled trials as a basis for developing treatment guidelines. Evid Based Med Health. 2018;21(1):4–6.

23.

Dunn

, Kaempf

, de Klerk

, et al. Vermont Oxford Network DRM Study Group: Randomized trial comparing 3 approaches to the initial respiratory management of preterm neonates. Pediatrics. 2011; 128(5):e1069–76.

24.

Vanpée

, Walfridsson-Schultz

, Katz-Salamon

, Zupancic

, Pursley

, Jónsson

. Resuscitation and ventilation strategies for extremely preterm infants: A comparison study between two neonatal centres in Boston and Stockholm. Acta Paediatr. 2007; 96(1):10–6; discussion 8-9.

25.

Akram Khan

, Kuzma-O’Reilly

, Brodsky

, Bhandari

. Site-specific characteristics of infants developing bronchopulmonary dysplasia. J Perinatol. 2006; 26(7):428–35. Epub 2006 May 18. Erratum in: J Perinatol. 2006;26(8):524.