Methods/Design
Trial design
Multicentre, randomised, masked, parallel controlled trial.
Participating centres
The following centres are actively recruiting for the trial:
Australia: Royal Hobart Hospital, Hobart; Royal Women’s Hospital, Melbourne; Monash Medical Centre, Melbourne; Women’s and Children’s Hospital Adelaide.
New Zealand: Auckland City Hospital, Auckland; Middlemore Hospital, Auckland.
United States: Evanston Hospital, Evanston, IL.
Turkey: Zekai Tahir Burak Hospital, Ankara.
The following centres have committed to joining the trial:
Australia: Mercy Hospital for Women, Melbourne.
New Zealand: Dunedin Hospital, Dunedin.
Israel: Ziv Medical Center, Tsfat; Bnai Zion Medical Center, Haifa.
United Kingdom: Southampton University Hospital, Southampton; Southern General Hospital, Glasgow; Royal United Hospital, Bath; University Hospital of Wales, Cardiff.
The Netherlands: University Medical Center, Groningen.
Poland: Poznań University of Medical Sciences, Poznań; Medical University of Łódź, Łódź;
Polish Mothers’ Memorial Hospital, Łódź; SPZOZ Provincial Hospital, Bydgoszcz.
Italy: San Gerardo Hospital, Monza; Ospedale Maggiore Policlinico, Milano.
Slovenia: University Medical Centre, Ljubljana.
Greece: Aristotle University of Thessaloniki, Thessaloniki.
Turkey: Uludag University Hospital, Bursa.
United States: Yale-New Haven Children’s Hospital, New Haven, CT; Children’s Hospital of Georgia, Augusta, GA; Cooper University Hospital, Camden, NJ; Kapiolani Medical Center, Honolulu, HI; Fletcher Allen Health Care, Burlington, VT; Beth Israel Deaconess Medical Centre, Boston, MA; West Virginia Health Science Center, Morgantown, WV; University of Southern California, Los Angeles, CA; Oklahoma University Health Science Center, Oklahoma, OK; Westchester Medical Center, Valhalla, NY.
Study population
Preterm infants of gestation 25 weeks 0 days to 28 weeks 6 days who are inborn and admitted to the NICU of a participating study centre, and who fulfil the entry criteria detailed below.
Recruitment
Entry criteria
1.
Requiring CPAP or nasal intermittent positive pressure ventilation (NIPPV) because of respiratory distress.
2.
CPAP pressure of 5–8 cm H2O and FiO2 ≥ 0.30.
3.
Less than 6 hours of age.
4.
Agreement of the Treating Physician in charge of the infant’s care.
Exclusion criteria
1.
Previously intubated, or in imminent need of intubation because of respiratory distress, apnoea or persistent acidosis.
2.
Congenital anomaly or condition that might adversely affect breathing.
3.
Identifiable alternative cause for respiratory distress (e.g. congenital pneumonia or pulmonary hypoplasia).
4.
Lack of availability of an OPTIMIST treatment team.
Consent
Written parental consent must be obtained prior to randomisation by the treating clinicians. A plain language document outlining the rationale for the study is given to the parents. Consent should be obtained prenatally where possible, in which case the infant will only be enrolled after birth if all inclusion and no exclusion criteria were fulfilled. In all cases, written consent is obtained using a specifically-designed consent form.
Randomisation
Once consent has been obtained and all entry criteria are met with no exclusions, the infant is randomised by the OPTIMIST Treatment Team, after handover of care from the treating clinicians. Enrolled infants are randomised into “surfactant via MIST” and “standard care” (sham treatment) groups, with an allocation ratio of 1:1, using a web-based randomisation procedure that requires confirmation of eligibility criteria and consent before revealing the randomly determined allocation. The randomisation is in randomly permuted blocks of variable length, stratified by study centre, and by gestational age. For the OPTIMIST-A trial there are two gestational age strata (25–26 weeks and 27–28 weeks). Twins and higher order multiples are randomised independently. Infants who are unstable and in need of immediate intubation should not be randomised, even if consent has been obtained; such infants will not be considered to have been enrolled.
Masking
In order to mask the group allocation from the treating clinicians, an OPTIMIST Treatment Team is mobilised to perform the randomisation and intervention. This team consists of a neonatologist, senior neonatal trainee or neonatal nurse practitioner, and a neonatal nurse, none of whom are currently involved in the infant’s care. Their role is to obtain the randomisation, and then within 1 hour, after screening the infant as effectively as possible from the treating clinicians, to administer the intervention (surfactant via MIST or sham treatment) in accordance with the randomised allocation. Their activities, including removal of surfactant from the medication refrigerator, movement and speech within the screened space, and manipulation of the infant, should be such that the treating clinicians cannot discern which intervention is received. All treating clinicians are made aware that the OPTIMIST Treatment Team will be concealing treatment allocation by performing a sham procedure on those infants randomised to standard care. The time taken to perform the intervention should be the same regardless of treatment allocation, and the infant is returned to the pre-intervention CPAP settings prior to removing the screens. A survey of clinical staff is being conducted after each OPTIMIST intervention in order to assess the success of masking.
Members of OPTIMIST Treatment Teams at all institutions undertake not to reveal the allocation group of randomised infants.
Intervention
The intervention is performed in the NICU of participating centres. Prior to intervention, all neonates must be stable on CPAP delivered by prongs or mask. An intravenous cannula should be in situ. It is desirable that a blood gas analysis (arterial or capillary) is performed before intervention, although this is not mandatory. A chest X-ray is recommended to confirm the diagnosis of RDS, and to exclude other causes of respiratory distress.
Having been briefed on the current condition of the infant, the OPTIMIST Treatment Team screens the infant from treating clinicians as completely as possible. The infant is then randomised, and the allocated intervention carried out as soon as possible (maximum 1 h after randomisation). Pre-intervention observations are recorded. The Treatment Team takes a labelled box containing full or empty surfactant vials from the OPTIMIST canister in the medication refrigerator. This canister must not be accessed by any other person other than the NICU pharmacist responsible for replenishing the stock of surfactant, which is supplied specifically for the study.
Intervention – surfactant administration via MIST
The following protocol is used for performing MIST:
Preparation
1.
Prepare the 16G Angiocath by marking a point indicating the desired depth of insertion beyond the vocal cords with a marker pen. The required depth is as follows: 25–26 weeks: 1.5 cm; 27–28 weeks 2.0 cm. Some investigators may find that tracheal catheterisation is facilitated by fashioning a slight anterior curve in the catheter.
2.
Draw up the surfactant (Curosurf™, Chiesi Farmaceutici, Parma, Italy) in a 3 or 5 mL syringe. The surfactant dose is 200 mg/kg (2.5 mL/kg). Draw up an additional 0.5 mL of air into the syringe, taking account of the dead volume of the instillation catheter (~0.3 mL).
3.
Optional: administer atropine 20 μg/kg intravenously.
4.
Disconnect standard monitors and connect the infant to the OPTIMIST oximeter (supplied to each centre)
5.
The infant can be swaddled and oral sucrose administered as part of standard procedural nursing care.
1.
Position the infant as for a standard intubation procedure.
2.
If possible, the laryngoscopy and tracheal cannulation should be performed with the CPAP prongs remaining in situ. An alternative which may improve the view of the vocal cords is to remove the CPAP prongs and apply CPAP by mask until the laryngoscopy commences.
3.
Perform direct laryngoscopy using a standard laryngoscope and blade. Alternatively, use the Glidescope Cobalt AVL video laryngoscope and size 0 stat.
4.
Insert the surfactant instillation catheter through the vocal cords to the desired depth, and hold it in position at the lips. The laryngoscope should then be removed.
5.
Connect the surfactant syringe to the catheter hub, and instil the surfactant in 2–4 boluses over 15–30 seconds.
6.
Once the surfactant is instilled, immediately remove the instillation catheter and apply CPAP by prongs or facemask.
7.
If on the first attempt catheterisation of the trachea is not possible within 20–30 seconds, remove the laryngoscope, allow recovery on CPAP as required, and then attempt tracheal catheterisation once again. The maximum number of catheterisation attempts should be 3, after which the procedure should be abandoned.
After MIST
1.
Once heart rate, SpO2 and respiratory effort are close to baseline values, restore the infant to their previous position, and re-establish CPAP with the same device and settings as prior to surfactant instillation.
2.
Details of the procedure are recorded on a data form specifically related to the intervention. This form is then removed from the bedside by the Treatment Team, and sent to the OPTIMIST Data Management Centre in electronic format. A copy of the form should be placed in locked cabinet away from the clinical area.
3.
Observations are recorded 5 minutes post-intervention, after which the OPTIMIST oximeter is disconnected and normal monitoring resumed.
4.
All items that could reveal the treatment allocation to the treating clinicians should be cleared from the bedside.
Because of the novelty of the MIST technique, OPTIMIST Investigators are given the opportunity to practise the technique on an intubation mannequin during an OPTIMIST training workshop. Experience from the feasibility studies at RHH Hobart and RWH Melbourne indicates that neonatologists and neonatal fellows are highly likely to succeed in tracheal catheterisation from the outset, although two attempts at catheterisation may be required until familiarity with the technique is gained.
Intervention – standard care (sham MIST procedure)
The following protocol is used in the standard care group:
1.
Position the infant as for a standard intubation procedure. This is the only actual intervention for babies randomised to standard care. CPAP is not interrupted at any time in this group. The OPTIMIST oximeter is used and standard monitoring disconnected as for the MIST procedure.
2.
Simulate the MIST procedure in terms of time taken and movement and communication within the screened area.
3.
After the procedure, restore the infant to their previous position, and ensure the CPAP settings are the same as prior to the sham procedure.
4.
Record the time of the sham intervention on the OPTIMIST Intervention Form, remove the form from the bedside, and send to the Data Management Centre, retaining a copy, exactly as described for the MIST procedure above.
Once the MIST procedure or sham procedure is completed, the screens around the infant are removed, and care of the infant returned to the treating clinicians. For all infants, attention is drawn to the possible need to reduce the FiO2 so as to keep SpO2 in the target range. An entry is made on the drug chart to indicate the timing of the OPTIMIST study intervention. A card is placed at the bedside indicating that the infant has been enrolled in the trial, and displaying the intubation criteria. Post-intervention observations are recorded by the treating clinicians at 4 hours.
Post-MIST investigations
A blood gas analysis (arterial or capillary) should be performed at 4 hours post-intervention, or earlier if clinically indicated.
Post-intervention management
Other than the requirement to adhere to intubation criteria in the first week, and in some cases perform a room air trial at 36 weeks corrected gestation, management of enrolled infants after intervention is at the discretion of the clinical team. Titration of CPAP pressure according to work of breathing and oxygen requirement is encouraged. Maximum acceptable CPAP pressure is 8 cm H
2O. NIPPV (bi-level CPAP) is allowable. Adjustment of FiO
2 should be so as to target an SpO
2 range appropriate for gestation and post-natal age. Prophylactic caffeine therapy would be expected in all infants [
41].
Criteria for intubation
Infants should be intubated and ventilated if, and only if, they fulfil any of the following criteria:
1.
FiO2 ≥ 0.45. To qualify for intubation, the FiO2 must be sustained at intubation level for at least 15 minutes, and all other aspects of CPAP management must have been optimised (including prong size and position, and minimisation of CPAP pressure leak).
2.
Apnoea unresponsive to caffeine therapy and stimulation, which is either frequent (6 episodes in 6 hours requiring vigorous stimulation), or severe (more than one episode requiring positive pressure ventilation)
3.
Persistent respiratory acidosis with pH < 7.20 and PCO2 > 65 mm Hg on two blood gas samples at least 30 minutes apart, or metabolic acidosis refractory to treatment
4.
Need for an anaesthetic or an intervention requiring intubation
Note that these criteria apply only during the first week of life, and only for the first episode of intubation.
Once intubated, surfactant therapy can be given, at the discretion of the treating clinicians. There is no likelihood of harm if a further dose of surfactant is given less than 6 hours after surfactant administration via MIST. Thus the treating clinicians remain masked in this circumstance.
Assessment of BPD at 36 weeks corrected gestational age
Incidence of BPD based on oxygen requirement at 36 weeks corrected gestational age is variable within units in the Australian and New Zealand Neonatal Network (ANZNN), certainly in part due to variability in approach to oxygen therapy amongst units. Given the primacy of BPD as an outcome in the OPTIMIST-A trial, a standardised approach to its recognition has been incorporated into the trial design, based around the National Institute of Child Health and Disease consensus panel definition of “physiological BPD” [
42]. On or shortly after 36 weeks 0 days corrected gestation, infants not requiring respiratory support (intubation/CPAP/HFNC ≥ 2 L/min) but receiving oxygen therapy with an FiO
2 of less than 0.30 have a trial of room air. For infants on nasal cannula oxygen the “effective FiO
2” is determined using the Benaron-Benitz formula [
43], currently available as on online calculator (
http://pub.emmes.com/study/rop/stop-js.html). Those with an FiO
2 < 0.30 have an air trial involving stepwise FiO
2 reductions 5 minutes apart until either room air is being administered or SpO
2 is no longer within the target range. Based on current evidence, the minimum acceptable SpO
2 reading for this trial is 91% [
44]. A successful room air trial is defined as SpO
2 readings ≥91% for 30 minutes in room air with nasal prongs removed [
42]. Oxygen therapy can thereafter be reinstituted if deemed necessary by the treating clinicians.
Infants receiving HFNC therapy with FiO2 0.21 and flow < 2 L/min also have a room air trial as above with the nasal prongs removed.
Infants requiring respiratory support, and those failing the room air trial, are deemed to have physiological BPD. BPD using the standard (clinical) definition is diagnosed if oxygen and/or respiratory support (intubation/CPAP/HFNC ≥2 L/min) is being administered for any portion of the day at 36 weeks and 0 days corrected gestational age.
Severity of BPD is categorised according to the consensus definitions [
42]:
Mild BPD: need for oxygen at 28 days but not at 36 weeks corrected gestation.
Moderate BPD: need for oxygen at 28 days and continued oxygen requirement at 36 weeks (confirmed by room air trial), with FiO2 < 0.30.
Severe BPD: need for oxygen at 28 days, and at 36 weeks an oxygen requirement with FiO2 ≥ 0.30 and/or need for positive pressure support (intubation, CPAP, HFNC ≥2 L/min).
Data collection and management
Within the OPTIMIST Investigator Team at each site, nominated personnel (e.g. Unit Data Collectors, Unit Research Nurses) collect data and enter it onto hard copy and/or electronic forms, as available. Data management is coordinated from the Clinical Epidemiology and Biostatistics Unit (CEBU) at the Murdoch Childrens Research Institute (MCRI), using a web-based database management system.
Data collection in hospital
Basic demographic, perinatal, and clinical data, as well as in-hospital outcomes, are collected prospectively for each patient, starting at enrolment. The data are entered on a hard copy clinical report form. Data pertaining to the MIST procedure are collected by the OPTIMIST Treatment Team, on a separate randomisation and intervention form. This form is not seen by other clinical or research staff. Once filled in, it is sent electronically to CEBU at MCRI. Information recorded includes the number of attempts required to catheterise the trachea, the total time taken, the lowest heart rate noted during the MIST procedure and time for restoration of heart rate above 100 beats per minute, the lowest SpO2 noted and time for restoration of SpO2 above 80%, and the need for and duration of positive pressure inflations by mask.
Data on heart rate, CPAP pressure, FiO2, and SpO2 prior to, and at four hours after intervention are collected, along with the results of pre- and post-intervention blood gas analysis.
Follow up
Each infant will have a full clinical and neurological assessment performed at 2 years by a developmental paediatrician blinded to the initial randomisation. Psychometric testing will be performed by a trained practitioner using the Bayley III Scales of Infant Development (or equivalent). Hospitalisation history in the first two years will be documented at this visit.
Outcome variables
Primary outcome
Incidence of composite outcome of death or physiological BPD [
42].
Secondary outcomes
A range of standard clinical outcomes pertaining to the first hospitalisation are being ascertained in trial participants. These are shown in Table
1. Additionally, data are being collected in the intervention group relating to the applicability and safety of the Hobart method (Table
2). For this purpose, data from the study oximeter are used alongside data recorded manually by the Treatment Team. Finally, longer term outcomes are being evaluated at two years corrected age as part of the OPTIMIST-A follow-up study. This study will have its own protocol and funding stream. Selected outcomes from this study are shown in Table
3.
Table 1
Clinical outcomes during first hospitalisation
| Duration of intubation (all episodes) |
Clinical BPD (oxygen or positive pressure support at 36 weeks corrected gestation) [ 45] | Duration of CPAP/NIPPV (all episodes) |
Mild/moderate/severe BPD [ 45] | Duration of intubation and CPAP |
Death | Duration of HFNC, minimum flow rate 2 L/min |
Death or BPD (clinical definition) | Duration of respiratory support |
Intraventricular haemorrhage (IVH) (all grades) | Duration of oxygen therapy |
IVH grades III and IV [ 46] | Requirement for oxygen at home |
Cystic periventricular leukomalacia | Length of stay in intensive care |
Retinopathy of prematurity (ROP) > stage II | Length of hospital stay |
Major morbidity (any of IVH grade III or IV, periventricular leukomalacia, ROP > stage II, physiological BPD) [ 47] | Total hospital billings |
Death or major morbidity | Calculated cost of hospitalisation |
NEC (Modified Bell stage 2 or greater) [ 48] | Pneumothorax requiring drainage |
NEC or spontaneous intestinal perforation requiring surgery | Pulmonary haemorrhage |
Requirement for intubation < 72 h | Patent ductus arteriosus (PDA) requiring anti-prostaglandin therapy |
Requirement for intubation at any time | PDA requiring ligation |
Need for additional surfactant therapy | Late onset sepsis (positive bacterial or fungal culture from a normally sterile site) |
Overall number of surfactant doses (including that given by MIST) | Time to regain birth weight |
Table 2
Applicability and safety outcomes in infants randomised to receive surfactant via the Hobart method
Incidence of successful surfactant administration via MIST | Duration of hypoxaemia (SpO2 < 80%) |
Number of catheterisation attempts | Requirement for, and duration of, positive pressure ventilation by mask |
Duration of bradycardia (heart rate < 100 beats per minute) | Incidence of apparent discomfort |
Table 3
Selected outcomes from the OPTIMIST-A follow-up study
Number of hospitalisations in the first 2 years | Major disability at 2 years |
Number of hospitalisations with respiratory illness in the first 2 years | Death or major disability at 2 years |
Statistical analysis and reporting
Statistical analysis
Data handling, verification and analysis for the OPTIMIST-A trial are being performed by CEBU at MCRI. Statistical analysis will follow standard methods for randomised trials and the primary analysis will be by intention to treat. For dichotomous outcomes, including the primary outcome in OPTIMIST-A, proportions will be compared using the odds ratio with 95% CI, obtained from a logistic regression analysis with adjustment for the strata (defined by centre and gestational age category) used in the randomisation. Continuous outcomes will be compared using differences between mean values, estimated from normal linear regression models with the same stratification adjustments. Secondary analyses will use expanded regression models to explore potential confounding effects of chance imbalances between arms in birth weight, gender, antenatal steroids, or mode of delivery. In further secondary analysis, we will explore evidence for heterogeneity of effects between the two gestational age strata in the trial, using interaction tests and subgroup analyses.
Data reporting and manuscript preparation
A clinical study report will be generated from the Data Management Centre. This document will, after approval by the Trial Steering Committee, form the basis of conference presentations and manuscripts for publication. In all cases data reporting will adhere to the CONSORT guidelines. Responsibility for manuscript preparation will rest with the Trial Steering Committee. Authorship will be in the form of: Author A, Author B, Author C,… for the OPTIMIST-A Investigators.
Sample size
In the RHH-RWH CPAP study, amongst infants of 25–28 weeks gestation the proportion positive for the outcome of death or BPD was 53% in those failing CPAP and 38% in those reaching the OPTIMIST-A enrolment threshold in the first 2 hours. A reduction by one-third in the proportion of infants with this outcome (i.e. from 38% to 25%) would be a major advance in care for this patient group, relieving the burden at both individual and NICU levels. Detection of a reduction of this magnitude with 90% power and α = 0.05 (two-sided) would require 297 subjects per arm [
49]. An allowance has been made for withdrawal of 2% of subjects post-recruitment. The number of subjects to be randomised in each arm is thus 303, for an overall total of 606.
Trial plan
Australasian neonatal Units and selected international centres, including those in the Vermont-Oxford network, are being invited to join the trial, and local information sessions held in interested centres as required. The Trial Coordinating Centre Team assists Units joining the study with ethics submissions and organisational matters.
At the time of study commencement at each site, a trial workshop is being conducted by a team from the Trial Coordinating Centre. These workshops consist of 1) a formal outline of the trial, 2) a hands-on demonstration of the MIST technique using an intubation mannequin, 3) a bedside simulation of the MIST procedure and of the sham intervention by an OPTIMIST Treatment Team, and 4) in-depth discussion of the practicalities of screening, randomisation and data collection.
A full complement of participating centres is expected for the OPTIMIST-A trial by mid-2015. Recruitment will thereafter proceed at full rate until completion, which is estimated to be completed at the end of 2017.
Data and safety monitoring committee
An independent Data and Safety Monitoring Committee (DSMC) has been established for the OPTIMIST-A trial. The terms of reference for this committee includes performance of interim data analysis, periodic examination of emerging external evidence in relation to MIST, and monitoring of adverse events, compliance with the trial protocol, and progress of recruitment. The DSMC has developed their own charter for the conduct of these and other activities [
50].
Interim analyses
The Trial Steering Committee expects that there will be a maximum of three interim analyses for the OPTIMIST-A trial. Relevant event rates in enrolled infants will be compared to the background rates in data from the RHH-RWH preterm CPAP study, and recommendations for change in sample size made if a substantial disparity is noted. The odds ratios for major outcomes will be examined in the two randomisation groups. For this comparison the statistical approach will be conservative, with a recommendation to cease the trial on efficacy grounds only to be made in the presence of very strong interim evidence. Later analyses will also include consideration of whether there is an unequivocal lack of efficacy. At each meeting of the DSMC the ethical position in relation to further randomisation will be considered based on results of other randomised controlled trials comparing MIST with standard care (continuation of CPAP).
Adverse events
Serious adverse events (SAEs), including those leading to death, prolonged hospitalisation or persistent disability, are relatively common in preterm infants 25–28 weeks gestation. Serious adverse events which are in the opinion of the local investigator unexpected, are to be reported within 5 working days to the coordinating centre. The SAE will then be reported to the local Ethics Committee, the Tasmania Health and Medical Ethics Committee, the Trial Steering Committee, the DSMC, and, as appropriate, the Therapeutic Goods Administration and/or other federal regulatory bodies. The data pertaining to the SAE will be examined by the DSMC, and any recommendations made will be disseminated to local investigators.
Funding
Funding has been obtained for commencement of the OPTIMIST-A trial from the RHH Research Foundation (for 2011 and 2012), and from the Australian NH&MRC for the years 2013–2017. Chiesi Farmaceutici, Parma, Italy has agreed to supply the surfactant at significantly reduced cost. The trial sponsor is the Menzies Research Institute Tasmania, and the trial insurer is Unimutual Limited.
PAD MBBS FRACP MD is a Neonatologist and Director of the Neonatal & Paediatric Intensive Care Unit, Royal Hobart Hospital, Hobart, and an Honorary Research Fellow at the Menzies Research Institute Tasmania.
COFK FRACP DMedSci is a Neonatologist at the Royal Women’s Hospital, Melbourne.
AGdeP FRACP MD is a Neonatologist at the Neonatal & Paediatric Intensive Care Unit, Royal Hobart Hospital, Hobart.
JBC BSc(Hons) PhD is Head of the Clinical Epidemiology and Biostatistics Unit, Murdoch Childrens Research Institute, University of Melbourne, Melbourne.
FO BSc MSc is a statistician at the Clinical Epidemiology and Biostatistics Unit, Murdoch Childrens Research Institute, Melbourne.
RFS MD PhD is H. Wallace Professor of Neonatology at the University of Vermont.
PGD FRACP MD is Professor of Neonatal Medicine at the Royal Women’s Hospital, Melbourne.
Competing interests
Surfactant for the intervention in the OPTIMIST-A trial is being provided at reduced cost by Chiesi Farmaceutici. The authors declare no other financial or non-financial competing interests.
Authors’ contributions
PAD: Conceived the study, made intellectual contribution to the study protocol, wrote the first draft of the manuscript. OK/AGdeP/JBC/FO/RFS/PGD: Made intellectual contribution to the study protocol, read and edited the manuscript. All authors read and approved the final manuscript. The authors of this manuscript constitute the membership of the OPTIMIST-A Trial Steering Committee.