Article Text

Download PDFPDF

Length of time from extubation to cardiorespiratory death in neonatal intensive care patients and assessment of suitability for organ donation
  1. S Saha1,
  2. A L Kent2,3
  1. 1Department of Obstetrics and Gynaecology, Canberra Hospital, Canberra, Australian Capital Territory, Australia
  2. 2Department of Neonatology, Canberra Hospital, Canberra, Australian Capital Territory, Australia
  3. 3Australian National University Medical School, Canberra, Australian Capital Territory, Australia
  1. Correspondence to Associate Professor Alison Kent, Department of Neonatology, Centenary Hospital for Women and Children, Canberra Hospital, PO Box 11, Woden, ACT 2606, Australia; alison.kent{at}


Objective A common concern for parents when end of life decisions are made is the length of time their baby may take to die. Postcardiac death organ donation is now becoming more common, along with neonatal organ donation. The aim was to determine the length of time from extubation until cardiorespiratory death (CRD) in neonatal intensive care patients and consideration of potential organ donation.

Design Retrospective review of medical records of neonates who died in a neonatal intensive care unit between 2000 and 2009.

Patients Data collected included gestation at birth, age at death, birth weight, reason for cessation of intensive care, inotrope and ventilation requirements, sedation and muscle relaxation prior to death, time from extubation to documented CRD. An assessment was made for potential suitability for consideration of organ donation with a gestation at birth ≥34 weeks and birth weight >2.0 kg.

Results 117 neonates were included, median gestation 29 weeks and median birth weight 1220 grams. The median age at death was 4 days of age. The median time from discussing prognosis to death was 137 min. The median time from extubation to CRD was 30 min. Seven (6%) neonates were considered suitable for organ donation, and for these infants the median time from extubation to CRD was 120 min. Two neonates donated heart valves.

Conclusions This provides a guide for grieving parents on time frames for the interval between extubation and CRD. More accurate postextubation CRD times are required to determine likely potential for postcardiac death organ donation.

  • Neonate
  • Organ Donation
  • Post Cardiac Death
  • Mortality

Statistics from

What is already known on this topic

  • Many neonates die in the neonatal intensive care after withdrawal of intensive care treatment.

  • There is little data in the literature for parents on how long neonates may take to die following extubation.

  • The requirement for organ donation in the paediatric population continues to be unmet.

What this study adds

  • The length of time that neonates of differing gestations and reasons for withdrawal of intensive care take to die following extubation.

  • A proportion of neonates may be suitable for postcardiac death organ donation.

  • Further research is required to determine more accurate times of death following extubation where organ donation may be considered.


The number of admissions of immature and critically ill infants to the neonatal intensive care units (NICUs) has increased significantly during the past three decades.1 Although survival rates have improved dramatically, in some instances, it is clear that prolongation of aggressive invasive treatment may not be in the best interests of the neonate. Studies in the past have shown that between 40% and 93% of deaths in NICUs are due to withdrawal of life-sustaining treatment.2–6

Following the parental decision of withdrawal of cardiorespiratory care, management is focused towards facilitating a death that is dignified and without suffering.7 Principles and practice of withdrawing life-sustaining treatments have been suggested when managing neonates and their families when a palliative plan has been decided.8 ,9 This may be a very distressing experience for parents and requires evidence to be able to provide parents with information on how long neonates may survive following the option of palliative care.

As with adults, children continue to die waiting for organ transplantation. Although children only form 1.5–2% of candidates on the active transplant waiting list, because of their size and improved organ support the number of children who could benefit from solid organ transplantation is increasing.10–12 In the UK only 10–20% of potential donation after brain stem death and donation after cardiorespiratory death's (CRD) postcardiac death organ donation (DCD) occur.13 There is now increasing evidence that paediatric DCD results in successful transplants.14–17 The important factor that must be considered in DCD is functional warm ischaemia time which is the time where the systolic blood pressure is less than 50 mm Hg in an adult (currently not defined in a neonate) and the time that cold perfusion begins at the time of surgery for organ extraction. The current accepted time in adults from withdrawal of life support and asystole for the following organs is 60 min for liver, 60 min for pancreas and 120 min for kidney.18 There are differences in times for lungs as time to reinflation is critical rather than time to cold perfusion.18 Due to the difficulty in diagnosing brain death in a neonate,19 ,20 there is little experience in solid organ transplantation from neonates. As infants less than 1-year-old on the heart transplant list have the highest mortality, Mathur et al21 have reviewed potential heart donors among newborns following CRD, finding that 4% of babies would have been potentially suitable. Along with organ donation, tissue donation can be considered including cardiac valves, bone, tendon and skin. Corneal donation from neonates is generally not included as the tissue is often too soft.

The aims of this study were to: (1) determine the time taken from extubation to CRD to provide a guide for clinicians and parents and (2) to determine the length of time to CRD in those neonates who might have been considered for DCD.


All neonates who died in the NICU between January 2000 and December 2009 were identified and a retrospective review of the medical records was performed to determine the length of time from extubation or withdrawal of respiratory support (continuous positive airway pressure (CPAP)) to CRD. Neonatal deaths that occurred in the delivery suite were not included. Data collected included: gestational age; birth weight; age at death; Clinical Risk Index for Babies-II (CRIB-II) score; inotropes, muscle relaxant, sedation, oxygen requirement, ventilation parameters, mean blood pressure and heart rate prior to withdrawal of cardiorespiratory support; type of sedative used; presence of raised creatine; coagulopathy; reason for palliative care; time from discussion of palliative care to CRD; time from extubation or withdrawal of CPAP to documentation of CRD; consideration for organ donation; donated organ; assessment as potentially suitable for organ donation.

The CRIB-II was calculated according to published methods.22 Chronic lung disease was defined as the need for supplementary oxygen and/or ventilator support at 36 weeks post the conception age.23 Intraventricular haemorrhage grading was based on Papile classification.24 Necrotising enterocolitis was staged using the modified Bell's Staging Criteria.25

Hypoxic-ischaemic encephalopathy (HIE) was diagnosed as per the Sarnat staging system.26

Persistent pulmonary hypertension was diagnosed with an oxygenation index greater than 20 and requirement for inhaled nitric oxide therapy.

An assessment was made from reviewing the medical records as to whether the neonate may have been appropriate for DCD organ donation. Exclusion criteria for potential organ donation included:

  • gestational age <34 weeks, birth weight <2.0 kg

  • neonates with multiorgan failure from infection

  • neonates with chromosomal anomalies

  • not ventilated at the time of decision for palliative care

Statistical analyses

All analyses and data were maintained using Predictive Analytics SoftWare (PASW Statistics V.18.0.2; SPSS: An IBM Company, Chicago, Illinois, USA, 2010). Data are presented as number (%), mean or median (IQR). Mann-Whitney, analysis of variance (ANOVA), χ2 and Fisher's exact tests were used as appropriate. The level of statistical significance for all analyses was set at p<0.05 using two-tailed comparisons.


One hundred and seventeen infants were included in the study. The median gestation was 29 weeks (IQR 25; 36) and the median birth weight 1220 g (IQR 692; 2410). The median CRIB-II score for neonates less than 31 weeks gestation was 12 (IQR 8; 14.75). The median age at death was 4 days (IQR 1.5; 9.5). In 93 (85%) cases the time from extubation to documented CRD was available with the median time being 30 min (range 1–2880; IQR 13.5; 65.5). Of the eight neonates who were CPAP dependent the median time from withdrawal of respiratory support was 1370 min (range 53–2880; IQR 53.5; 2831).The median time from discussing prognosis to CRD in all cases was 137 min (IQR 55; 300).

Sixty-nine patients (59%) were on intermittent ventilation, 40 (34%) were on high frequency oscillatory ventilation and 8 (7%) were not ventilated. Sixty (55%) neonates were receiving inotropes at the time of extubation and 23 (21%) were receiving a muscle relaxant. Eighty-three (76%) were receiving some form of intravenous sedation prior to extubation. Sedation used was either morphine alone (52 neonates (63%)) or morphine and midazolam (33 neonates (37%)). Time of documented CRD from extubation was significantly shorter in those receiving inotropes and muscle relaxants (p=0.005 and 0.005, respectively). Receiving sedation was not associated with a shorter time of documented death from extubation (p=0.06). The median PIP prior to death was 23 cmH2O_ (IQR 18; 30.8) and positive inspiratory pressure (PEEP) 6 cmH2O (IQR 5.0; 7.0). The median mean blood pressure (BP) was 33.5 mm Hg (IQR 27; 44.3) and heart rate 135 beats per minute (IQR 102.5; 160). The mean blood pressure was less than the 5th centile in 36 cases (31%).

The causes of death are described in table 1. The mean gestation (figure 1) and birth weight at birth were significantly higher in those that died from HIE, and neuromusucular and other disorders (p<0.0001 and p<0.0001, respectively). There was a significant difference between the cause of death and length of time to documented CRD from extubation (p=0.03) (figure 2).

Table 1

Documented reason for withdrawal of intensive care

Figure 1

Cause of death by mean gestational age.

Figure 2

Cause of death and median time from extubation to cardiorespiratory death (CRD).

Two patients donated heart valves, for the remaining patients the option of donation was not documented in the medical records. One infant that donated heart valves was 28 weeks gestation, 1210 g at birth and died on the third postnatal day. Withdrawal of cardiorespiratory support was for Grade IV intraventricular haemorrhage and CRD was documented 90 min following withdrawal of support. The other infant was 37 weeks gestation, 2795 g at birth and died on the fourth postnatal day. Withdrawal of cardiorespiratory support was for HIE and CRD was documented 52 min following withdrawal of support. Following review, seven cases were considered potentially appropriate for organ donation. These babies had a gestation of ≥34 weeks gestation and birth weight >2.0 kg. The median time from extubation to documented CRD in these neonates was 120 min (IQR 52; 720). The length of time from extubation until documented CRD was significantly longer in this group of neonates (p=0.001). Of the seven cases considered suitable for organ donation only one was receiving inotropes at the time of extubation.


This study provides parents and clinicians with an estimation of the time of death following cessation of cardiorespiratory support of neonates by condition, gestational age and birth weight. This study also uniquely assesses time from extubation to documented CRD for neonates who may be considered for organ donation.

Our study is limited by the fact that it is retrospective in nature, does not have the time from extubation to CRD in all cases, cardiac monitoring was not performed at the time of withdrawal of intensive care and limited documentation regarding organ donation was available. Despite these limitations, this is the first study in Australia, and one of few worldwide where importance of documented time of CRD following extubation in neonates has been reviewed in the context of potential neonatal organ donation. Without proper documentation it is impossible to have an accurate picture of how many families could have potentially donated organs.

A study by McHaffie et al7 concluded that once the decision has been made for withdrawal of life-sustaining treatment parents expect their child to die soon without suffering. While medication such as sedation and muscle relaxants can be continued during withdrawal27 our study has shown that sedation prior to palliative care was not associated with a reduction of the time from extubation to death. This information provides reassurance for parents and clinicians that the use of sedation prior to palliative care does not accelerate the dying process.

The general practice in this unit was that all infusions including inotropes, muscle relaxants, sedation and analgesia were continued until the time of extubation. After extubation the families were in a private room without constant medical or nursing attendance or continuous cardiac monitoring with intermittent assessment as to when CRD had occurred. As such the documented times of death for this study are an overestimation of the time taken until CRD occurred, and as such should not be used as a guide to state that time taken to CRD in neonates would be too long to consider for DCD organ donation.

Janvier et al28 described the time to CRD from extubation between two groups as unstable (which included two of either persistent desaturations despite 100% oxygen on mechanical ventilation, hypotension despite volume infusion and inotropes or protracted bradycardia, or anuria for >24 h) or stable with the mean time to CRD in these two groups being 39 min and 357 min. If we look at our group receiving inotropes to compare against these findings the mean time to CRD for those receiving inotropes was 37 min and for those not on inotropes 166 min. Verhagen et al looked at the median time to CRD following extubation for those who were described as no chance to survive and those with a very poor prognosis with the times being 51 min and 38 min, respectively.29 Although we are unable to compare our results directly with this study the times are possibly similar and provide some guidance to clinicians and families on what to expect for neonates having withdrawal of cardiorespiratory support. Neither of these studies describe whether there was continued cardiac monitoring to determine the time of CRD or whether intermittent assessments were made.

In the USA there are over 1200 children on the waiting list for solid organs including kidney, heart, liver and lungs.11 Although the number of organ donations has increased over the years, there is a significant disparity between the number of patients requiring organs and the number of organs that are available. In the absence of absolute contraindications, it is reasonable to enquire as to any parent's wishes regarding organ donation. Successful cases are reported of DCD neonatal hearts being donated from babies in Europe and being transplanted in infants in the UK, and discussions are again occurring on the ethical considerations of anencephalic neonates donating organs. In our study the neonates considered suitable for organ donation, had a median length of time from extubation to CRD which would be considered too long for solid organ donation.30 However, as described earlier this is an overestimation of the time where CRD has occurred. In these circumstances families must be prepared for the possibility that if the child does not die within the allocated time following extubation, he or she cannot be a solid organ donor. However, tissue donation may still remain an option.

DCD organ donation times are based on adult studies. In our study the majority of patients considered suitable for organ donation did not require inotropic support, indicating that they had good cardiovascular function and that all organs would have been receiving good perfusion prior to extubation. In red cells of adults and children more than 6 months of age, haemoglobin A accounts for 95% of the total haemoglobin. On the other hand in the fetus, haemoglobin F (HbF) is the predominant haemoglobin from 8 weeks gestation with levels continuing to increase until 20–24 weeks gestation. Because of the unique properties of HbF, fetal red cells have a considerably higher oxygen affinity than do adult red blood cells.31 ,32 Physiologically, this property ensures that the haemoglobin of the fetus is oxygenated at the expense of maternal haemoglobin A, thereby facilitating the transport of oxygen across the placenta. The proportion of HbF decreases as gestational age increases. The concentration of HbF in an infant born at 28 weeks gestation is approximately 90%, and decreases to approximately 60% at 10 weeks after birth.33 Given the presence of HbF it is possible that the warm ischaemia time for a neonate may be longer for DCD organ donation than in an older child or adult.


This study provides estimated times of CRD following cessation of cardiorespiratory support of neonates by condition, gestational age and birth weight and considers the time to CRD in those who might be suitable for DCD organ donation. Further studies are required to accurately determine time to CRD in neonates and whether neonatal organs would have longer warm ischaemia times than adult organs due to HbF to ultimately increase the potential organs available for paediatric transplantation.


We thank the medical records department for their assistance with this study.



  • Contributors Both authors of this paper have: (1) made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafted the article or revised it critically for important intellectual content; and (3) approved the final version to be published.

  • Competing interests None.

  • Ethics approval This study was approved by the ACT Human Research Ethics Committee under the NHMRC guidelines for clinical audit.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

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.