Objectives Therapeutic hypothermia improves neurological outcome in adults after ventricular fibrillation cardiac arrest and neonates with hypoxic ischaemic encephalopathy. There is currently no clinical research to support its use in the paediatric population. This survey aims to ascertain current practice in the UK, and attitudes and opinions to guide the feasibility of a UK multicentre, randomised, controlled trial of therapeutic hypothermia after cardiac arrest in children (The Cold-PACK Post Arrest Cooling in Kids study).
Methods Anonymous survey of UK paediatric intensive care consultants (n=149).
Results A total of 113 (76%) of 149 surveys were returned; 65% responded that they do not know if therapeutic hypothermia improves survival after cardiac arrest. Despite this, 48% ‘always’ or ‘often’ use therapeutic hypothermia after return of spontaneous circulation following cardiac arrest in children. Among those who never use therapeutic hypothermia (33%) the commonest explanation given was ‘not enough research evidence’ (91%). With respect to the dose of therapeutic hypothermia the median duration of cooling used is 24–48 h (range 4–72 h) and median target temperature 34°C to 35°C (range 32°C to 37°C); 68% target a temperature range higher than that applied in the published adult and neonatal studies (33±1°C). There was strong support for a trial of therapeutic hypothermia being ethical (89%) and using deferred consent (85%).
Conclusions Wide variation in UK practice in the use of therapeutic hypothermia and a state of clinical equipoise is demonstrated by this survey, which shows important support for UK multicentre collaboration in a future trial of therapeutic hypothermia after cardiac arrest.
Statistics from Altmetric.com
Therapeutic hypothermia to 33±1°C instigated after return of spontaneous circulation (ROSC) in adults who remain comatose 1 h after ventricular fibrillation cardiac arrest has been shown to significantly improve neurological outcome.1 2 Four large trials have also demonstrated that 72 h of therapeutic hypothermia improves neurological outcome in newborns with hypoxic ischaemic encephalopathy.3,–,6
The use of therapeutic hypothermia after cardiac arrest in children has not been adequately studied to date. Despite this the International Liaison Committee for Resuscitation (ILCOR) recommend considering the use of therapeutic hypothermia for 12–24 h in children who remain comatose after resuscitation from paediatric cardiac arrest; presumably extrapolating from neonatal and adult literature.7
An anonymous web-based survey conducted in North America highlighted wide variation of practice in paediatric intensive care units concerning when therapeutic hypothermia is started and how it is administered.8 As this survey was undertaken 5 years ago, with little input from UK intensivists and a poor overall response rate (12%), it does not provide an accurate picture of current UK practice. This survey aims to ascertain current practice in the UK and attitudes and opinions to guide the feasibility of a UK multicentre, randomised, controlled trial of therapeutic hypothermia after cardiac arrest in children (The Cold-PACK Post Arrest Cooling in Kids study).
What is already known on this topic
▶. Therapeutic hypothermia is an established treatment for neonates with hypoxic ischaemic encephalopathy after birth and adults after ventricular fibrillation cardiac arrest.
▶. There is little published research on the use of therapeutic hypothermia after paediatric cardiac arrest and wide variation in its use.
What this study adds
▶. In the UK approximately half of paediatric intensive care consultants currently use therapeutic hypothermia after paediatric cardiac arrest.
▶. There is wide variation in the duration and depth of therapeutic hypothermia used.
▶. Support for future UK multicentre research is currently present.
Relevant questions were generated by the study group and piloted on a group of 10 paediatric intensive care (PIC) consultants from 3 hospitals for further feedback and hyperlink access testing.
In all, 149 PIC consultants from 25 UK Paediatric Intensive Care Units (PICUs) were identified. An invitation to participate was sent out on 1 October 2008. Links to the survey were sent out a further three times and the website was closed on the 23 November 2008.
This was created using Microsoft ASP.NET 2008 (Microsoft, Seattle, Washington, USA). Microsoft Excel (Microsoft, Seattle, Washington, USA) was used for data analysis and results are presented as percentage of survey responders, or mean (SD).
Of 149 consultants surveyed, 113 responses (76%) were received.
Consultants responded to questions about their current use of therapeutic hypothermia (table 1), how they select patients (table 1), what ‘dose’ of hypothermia they use (table 1), methods of cooling (table 2) and their views on further research (table 1).
Use of therapeutic hypothermia
A total of 65% reported not knowing if therapeutic hypothermia after cardiac arrest improves survival; 48% always or often use therapeutic hypothermia with the remaining 52% stating they seldom or never use it. The commonest reason given by those who seldom or never use hypothermia was ‘not enough research evidence’ (91%).
A total of 65% of PIC consultants who have used therapeutic hypothermia would not change their practice according to whether the cardiac arrest occurred in or out of hospital.
The time to ROSC after cardiac arrest was influential in decision making for 66% (table 1). A higher proportion would commence therapeutic hypothermia for ROSC times between 5 and 30 min, though 23% would even consider therapeutic hypothermia for a ROSC delay exceeding 60 min.
Methods of inducing and maintaining therapeutic hypothermia
A total of 90% of PIC consultants do not have a protocol for the use of therapeutic hypothermia in their unit.
Various combinations of cooling methods were reported, with a circulating water blanket the most commonly used (78%) (table 2).
‘Dose’ of hypothermia
The stated duration of induced hypothermia ranged from 4 to 96 h. The maximum duration of cooling used is shown in figure 1. The majority cool for a duration of up to 24–48 h (85%) with 15% having a maximum cooling time longer than 48 h.
Depth of cooling
This ranged from 32 to 37°C with only 33% targeting 33±1°C (figure 2). The two responders (3%) who stated 36°C to 37°C commented that they use cooling methods to actively avoid hyperthermia rather than to induce hypothermia.
Speed of rewarming
This was controlled by 50% of responders with a range of 0.1°C to 1.0°C/h given as the target temperature increase.
Views on future research into therapeutic hypothermia
There continues to be a state of clinical equipoise in the UK regarding the use of therapeutic hypothermia after cardiac arrest according to 73% of PIC consultants.
A total of 86% would randomise their patients into a randomised controlled trial (RCT) of therapeutic hypothermia in cardiac arrest. There was greatest support for a comparative trial of hypothermia versus normothermia. There was strong support for a RCT of therapeutic hypothermia being ethical (89%) and using deferred consent (85%).
Although 83% of those surveyed felt that publication of a large multicentre study undertaken in the USA would inform them of the benefit or lack of benefit of therapeutic hypothermia to their own patient population, 46% agree that a UK trial is needed in addition to a US trial.
This survey suggests that there is wide variation in the use of therapeutic hypothermia after cardiac arrest in children in PICUs in the UK. Approximately 50% of PIC consultants report ‘always’ or ‘often’ using it, despite an absence of published, paediatric specific, literature to support its use. There is however an ILCOR recommendation ‘to consider the use of therapeutic hypothermia after cardiac arrest in children’.7 The evidence in animal, neonatal and adult literature, although promising for those patient groups and specific conditions, has limitations when transferred to the paediatric cardiac arrest population.
Simple surface methods of cooling were used by all responders, with the majority reporting using more than one method. Of interest is the small number (5%) who reported to use 4°C iced intravenous saline to induce therapeutic hypothermia, a method demonstrated to be safe and effective in adult pilot studies of therapeutic hypothermia9,–,11 and one paediatric case report.12
The ‘dose’ of hypothermia appears to be an important element in the potential efficacy of therapeutic hypothermia. The wide variation in duration and depth of cooling and also rate of rewarming reflects the lack of clear evidence. Very few centres reported following a unit protocol that would assist in unifying centre practice. Uncertainty may also be present after the reporting of a trend towards increased harm with the use of hypothermia in the Hypothermia after Paediatric Head Injury (Hyp-HIT) trial.13
A total of 86% of PIC consultants stated they select a duration of cooling between 24 and 48 h although durations outside this range are also considered. The two adult therapeutic hypothermia RCTs used 12 or 24 h duration1 2 while the neonatal studies of hypoxic ischaemic encephalopathy have used 72 h duration.3,–,6 This survey demonstrates a stated duration of treatment falling between the two sets of published data. Without specific safety data in children the potential risks of longer duration of cooling are not known.
In all, 74% of PIC consultants feel there is still clinical equipoise for the use of therapeutic hypothermia after cardiac arrest in children. A further study,6 supporting therapeutic hypothermia for neonates with hypoxic–ischaemic encephalopathy, was published after the survey was performed. It is conceivable that publication of this and other relevant studies could have an impact on clinicians views and behaviour.
The recently commenced Therapeutic Hypothermia After Paediatric Cardiac Arrest (THAPCA) and Hypothermia after Cardiac Arrest in Paediatric (HypCAP) trials in North America and Canada will hopefully start to answer the question of safety and efficacy of this treatment in children. There is agreement (83%) that the results of these studies are likely to inform clinical practice in the UK, but support for a study also to be undertaken in the UK, which perhaps reflects that a number of trials showing an effect are required to change clinician's behaviour.
The use of deferred consent is a new area in paediatric resuscitation research in the UK. Recent changes in April 2008 to the Medicines for Human Use (Clinical Trials) Regulations, UK 2004 have allowed the use of deferred consent (with ethics committee approval) to be applied to emergency resuscitation research.14 In this survey 85% supported the use of deferred consent. The use of deferred consent may enable very earlier instigation of treatment during a clinical trial and therefore maximise the potential benefits of therapeutic hypothermia.
One of the strengths of this survey is the high response rate of 76%. A targeted approach to a defined population of UK PIC consultants with direct clinical involvement and decision making regarding patient care enable a degree of confidence that these findings reflect the current UK PIC community practice. As with all surveys however the findings only represent self-reported behaviour and do not necessarily equate to actual clinical behaviour. However, the findings give additional support for the need and willingness to undertake further investigative research into the use of therapeutic hypothermia after paediatric cardiac arrest.
The authors are indebted to the paediatric intensive care consultants in the UK who took the time to complete this survey and Mr David Scholefield for the web-based support and survey design.
Competing interests None.
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.