What is the effect of cardiopulmonary resuscitation at birth on extremely premature infants?
- Report By: Donovan Duffy - Neonatal Specialist Registrar
- Institution: St Georges Healthcare NHS Trust
- Date Submitted: 22nd July 2009
- Date Completed: 12th November 2010
- Last Modified: 12th November 2010
-
Status:
Green (complete)
Three Part Question
In [extremely premature infants] what is [the effect of CPR at delivery] on [survival and neurodevelopmental outcome]?Clinical Scenario
A premature baby born at 24 weeks gestational age is admitted to the neonatal unit having been born in poor condition and receiving cardio-pulmonary resuscitation (CPR) with adrenaline in the delivery room. Considering the available evidence, is the use of CPR at delivery of extremely premature infants associated with very poor outcomes such that CPR in these infants may be inappropriate? Does the administration of CPR provide these infants with a chance of survival free of disability?Search Strategy
1. Cochrane Library accessed via http://www.thecochranelibrary.com/2. MEDLINE via Pubmed (1950-present)accessed via http://www.ncbi.nlm.nih.gov/pubmed/
1. Cochrane Library: using "advanced search" with terms ‘premature infant’ AND ‘cardiopulmonary resuscitation’ entered in "search all text" in "all of the cochrane library".
2. Pubmed: Limits: All Infant: birth-23 months. ‘premature OR low birth weight infant AND cardiopulmonary resuscitation AND outcome’ ‘premature OR low birth weight infant AND adrenaline resuscitation AND outcome’.
Initial search conducted May 2008 and updated June 2009.
Search Outcome
1. Cochrane Library: One relevant review found but not considered as it provides no results- no randomised, controlled trials were found meeting criteria for inclusion in this Cochrane review.(1)2. Pubmed: ‘premature OR low birth weight infant AND cardiopulmonary resuscitation AND outcome’- 41 studies, 5 relevant.(2-6) ‘premature OR low birth weight infant AND adrenaline resuscitation AND outcome’- 22 studies, 2 further relevant articles found. (7,8) Two additional studies found cross referencing articles. (9,10) Review publications, studies describing cardiopulmonary resuscitation outside of delivery and articles unavailable in English were not considered. Another study was not included as it described the outcome for only one infant less than 28 weeks gestation who received CPR after adequate prior respiratory support. (11)
Relevant Paper(s)
| Author, date and country | Patient group | Study type (level of evidence) | Outcomes | Key results | Study Weaknesses |
|---|---|---|---|---|---|
| Davis (2) 1993 Canada | 156 infants <1000g. 500-750g: 8 received CPR. 751-1000g: 10 received CPR | Case series(level 4) | Survival | 500-750g: 8 recipients of CPR all died. 751-1000g: 3 recipients of CC’s all survived, 4 of 7 recipients of adrenaline & CC’s survived. | Gestational ages not described Apgar scores not specified for CPR recipients |
| Cranial US | Survivors: No IVH/ PVL on US. | ||||
| Neurodevelopment | 1-2 years corrected age- 4 normal, 2 mild motor delay. | ||||
| CPR rate (%) | 11.5 | ||||
| Survival rate to discharge for CPR recipients | 39% (7/18) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | 0% (0/7) | ||||
| Abnormal neurodevelopment for CPR survivors assessed | 29% (2/7) | ||||
| Survival with normal neurodevelopment for all CPR recipients | 22% (4/18) | ||||
| Finer (3) 1999 USA | 27707 infants 401–1500 g 401–500 g: 46 received CPR (5% received epinephrine) 501–750 g: 656 received CPR (8.6% epinephrine) 501–1500 g: 1618 received CPR (4% epinephrine) | Case-control study (level 4) | Survival | CPR vs No CPR: 401–500 g: 24% vs 16% (p=0.17). 501–750 g: 44% vs 62%. 501g-1500g: 63% vs 89% [p<0.001]. | 401–500 g: mean (SD) GA 23 (2) weeks 501–1500 g: mean (SD) GA 28 (3) weeks 88% of 501–1500 g CPR recipients had 1 min Apgar scores of 3 or less Large geographically defined cohort White matter injury not studied No neurodevelopmental follow-up |
| IVH | No IVH- 501g-1500g: 62% vs 79%. 501g-750g: 60% vs 66%. Survival without IVH grade 3/ 4- 501g- 1500g:53% vs 84.6% [p<0.01]. 501g-750g: 37% vs 54%. | ||||
| CPR rate (%) | 11 | ||||
| Survival rate to discharge for CPR recipients | 62% (1024/1618) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | 63% (501–750 g) | ||||
| Abnormal neurodevelopment for CPR survivors assessed | Not assessed | ||||
| Survival with normal neurodevelopment for all CPR recipients | Not assessed | ||||
| Finer (4) 1999 USA | 177 infants <1000 g 19 received CPR | Case-control series (level 4) | Survival | CPR vs No CPR: <1000 g: 79% vs 71%. <750 g: 77% vs 52% . | Mean (SD) GA 26 (1.4) weeks; range 24–28 weeks Median Apgar scores at 1 min 2 (range 0–5) All 4 CPR mortalities at 24 weeks gestation Neurodevelopmental follow-up to median of 28 months 5 Lost to follow-up, 1 died shortly after discharge Unclear whether these infants similarly described as part of Vermont Oxford cohort |
| Cranial US | IVH: 70% vs 10% [p=0.015]. Non significant differences: echolucencies, ventricular dilatation, white matter injury. 15 survivors: 7: normal cranial US. 2: grade 3 IVH/ echolucencies /ventricular dilatation. 13 recipients of CPR <750 g: 10 survivors: 1 CC, 2 CC and epinephrine, 7 only epinephrine. | ||||
| Neurodevelopment | 10 CPR survivors at >10 months 7 normal, 1 questionable, 2 abnormal (CP) | ||||
| CPR rate (%) | 11 | ||||
| Survival rate to discharge for all CPR recipients | 79% (15/19) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | 47% (501–1500 g) | ||||
| Abnormal neurodevelopment for CPR survivors assessed | 20% (2/10) | ||||
| Survival with normal neurodevelopment for all CPR recipients | 37% (7/19) | ||||
| Jankov (5) 2000 Canada | 198 infants < 750g. 16 received CPR, 12 of which received adrenaline (all via endotracheal tube) | Case-control study (level 4) | Survival | CPR vs no CPR: Non significant differences 56% vs 72%. | Median GA 25 weeks; range 23–33 weeks Median Apgar score at 1 min 1 (0–5) All 3 infants with grade 3/4 IVH or PVL died Neurodevelopmental follow-up to median of 24 months |
| Cranial US | IVH Grade 3/4: 13% vs 10%. PVL: 6.2% vs 6%. | ||||
| Neurodevelopment | Adverse neurodevelopment (non ambulation, quadriplegia, blind, hearing loss with aids, IQ 2 SD < mean): 11% vs 23%. | ||||
| CPR rate (%) | 5.7 | ||||
| Survival rate to discharge for CPR recipients | 56% (9/16) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | 0% (0/9) | ||||
| Abnormal neurodevelopment for CPR survivors assessment | 11% (1/9) | ||||
| Survival with normal neurodevelopment for all CPR recipients | (“free of severe disability” –50%; 8/16) | ||||
| Sanches-Torres et al (5) 2007 Spain | 150 infants 425–995 g 32 infants received CPR compared to 118 infants who did not receive CPR | Case-control study (level 4) | Survival | CPR vs no CPR: Non significant differences 62.5% vs 76.3%. | Mean (SD) GA 25 (1.2) weeks; range 23–27 weeks 31% of CPR infants had Apgar scores of 3–5 at 1 min No neurodevelopmental follow-up |
| Cranial US | IVH: 63% vs 53%. Cystic PVL: 16% vs 11%. HPI: 19% vs 11%. IVH grade 3: 31% vs 17%. IVH grade 3 and/or cystic PVL and/or HPI: 47% vs 22% [p=0.001]. CPR group: larger air leaks and coagulopathy (p<0.01) | ||||
| CPR rate (%) | 21 | ||||
| Survival rate to discharge for CPR recipients | 63% (20/32) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | (31% (10/32) for CPR recipients; unclear for survivors) | ||||
| Abnormal neurodevelopment for CPR survivors assessed | Not assessed | ||||
| Survival with normal neurodevelopment for all CPR recipients | Not assessed | ||||
| O’Donnell (6) 1998 Australia | 78 infants who received adrenaline at birth of which 23 infants were <28 weeks. (10 infants < 750g). | Case series(level 4) | Survival | 30% < 28 weeks. (50% < 750g). | Mean (SD) GA 26 (1.8) weeks; range 24–28 weeks Median Apgar score at 1 min 1 (IQR 1–3) Neurodevelopmental follow-up “to at least 1 year |
| IVH | IVH grade 3 or 4: 17%. | ||||
| Neurodevelopment | 71% normal, 29% severe disability. | ||||
| CPR rate (%) | Not described | ||||
| Survival rate to discharge for CPR recipients | 30% (7/23) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | (17% (4/23) for CPR recipients: unclear for survivors | ||||
| Abnormal neurodevelopment for CPR survivors assessed | 29% (2/7) | ||||
| Survival with normal neurodevelopment for all CPR recipients | 22% (5/23) | ||||
| Sims et al (8) 1994 UK | 105 infants (24–42 weeks GA) who received epinephrine and/or atropine as part of resuscitation at any time) 20 treated at birth: 5 babies 28 weeks or less | Case series(level 4) | Survival | 3 deaths | Mean (SD) GA 26 (1.8) weeks Apgar scores at 1 min of 0–4 CCs not studied 41% mortality for all births ≤28 weeks |
| Normal/ Handicap | 2 survivors: Both with cerebral palsy and left hemiplegia. | ||||
| CPR rate (%) | Not described | ||||
| Survival rate to discharge for CPR recipients | 40% (2/5) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | Not described | ||||
| Abnormal neurodevelopment for CPR survivors assessed | 100% (2/2) | ||||
| Survival with normal neurodevelopment for all CPR recipients | 0% (0/2) | ||||
| Francis (9) 1997 Canada | 163 infants <1250g. 36 received CC. 17 survivors of CC compared to 63 matched survivors who did not receive CC. | Case-control study(level 4) | Survival | CC vs no CC: 47% vs 76% [p<0.001]. | CPR rate high at 22%. Published in abstract form only. |
| IVH, neurodeveloment, hospital stay, seizures, hearing loss | Non significant differences in: Rates/ severity of IVH, seizures, hearing loss, CP, neurodevelopmental quotients. | ||||
| CPR rate (%) | 22 | ||||
| Survival rate to discharge for CPR recipients | 47% (17/36) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | Not described | ||||
| Abnormal neurodevelopment for CPR survivors assessed | Not described | ||||
| Survival with normal neurodevelopment for all CPR recipients | Not described | ||||
| Jain et al (10) 1991 USA | 93 infants with Apgar scores of 0 at 1 min and CPR at birth 750 g or less: 10 infants 751–1500 g: 19 infants | Case series(level 4) | Survival | All 10 < 750g died. 47% survival 751-1500g. | GA not described Survival of cohort (<750 g to >2500 g): 39% |
| CPR rate (%) | Not described | ||||
| Survival rate to discharge for CPR recipients | (0% (0/10): <750 g; 47% (9/19): 751–1500 g) | ||||
| Rate of IVH grade 3 or 4 for CPR survivors | Not described | ||||
| Abnormal neurodevelopment for CPR survivors assessed | Not assessed | ||||
| Survival with normal neurodevelopment for all CPR recipients | Not assessed |
Comment(s)
Current Resuscitation Council recommendations on the use of CPR and epinephrine in newborns do not take into consideration gestational age or birth weight.(International Liason Committee on Resuscitation)The British Association of Perinatal Medicine framework for practice on the management of babies born extremely preterm states: “There is no evidence to support the use of epinephrine by any route, or chest compressions, during resuscitation at gestational age <26 weeks”.(Wilkinson)
Evidence on outcome after delivery room CPR is limited to retrospective studies. Earlier small cohort studies reported unfavourable outcomes in all survivors (Sims) and 100% mortality in CPR recipients of less than 750 g.(Davis, Jain). These results, taken in the context of poorer outcomes for extreme prematurity, caused the authors of early studies to question the use of CPR at birth in extremely preterm infants (Davis, Sims). Advances in perinatal practice have since led to improved overall survival for infants born at 24 weeks gestation and above.(Costeloe)
Survival rates to discharge for extremely low birthweight infants given CPR at birth range from 44% to 79% in studies published since 1999.(Finer, Finer, Jankov, Sánchez-Torres ). When compared to non-CPR control groups, survival rates in CPR recipients were statistically significantly lower in Francis' cohort (<1250 g birth weight) and Finer's Vermont Oxford cohort (501–1500 g birth weight).(Finer, Francis). Survival for CPR recipients of birth weight 501–750 g in the Vermont Oxford cohort was lower but not significantly different.(Finer) Survival rates showed lower trends in two other studies.(Jankov, Sánchez-Torres)
CPR recipients in Finer's San Diego cohort and of birth weight 401–500 g in the Vermont Oxford cohort showed higher survival rates than controls. The authors stated that survival was higher in these CPR recipients probably because extremely small infants, those most premature or not expected to survive, were not resuscitated.(Finer, Finer)When survival rates between CPR recipients and controls are compared, selection bias could conversely push survival rates for CPR recipients lower when more of the smallest, most premature or non-viable infants are offered CPR. Infants who receive CPR are likely to have more adverse perinatal risk factors and are born in poorer condition compared to controls, thus pushing survival rates lower. All deliveries with signs of life (including those around the limits of viability) need to be recorded as live births to enable accurate calculation of outcomes.(Evans) We do not know the accuracy of recording of live births for these studies.)
Higher numbers of babies are required to accurately determine rates of abnormalities in survivors. Many CPR recipients survived with abnormalities comparable to controls. There were few statistically significant differences: survival without intraventricular haemorrhage (IVH) grade 3/4 in the Vermont Oxford cohort,(Finer) survival with IVH in the San Diego study (Finer) and composite survival with IVH grade 3 and/or cystic periventricular leucomalacia and/or haemorrhagic periventricular infarction in the Spanish cohort.(Sánchez-Torres). Three cohorts showed no increased IVH in CPR survivors.(Davis, Jankov, Sánchez-Torres). O'Donnell showed a 17% rate of grade 3/4 IVH in CPR recipients (no comparative control group) and two studies did not examine ultrasound outcomes.(O'Donnell, Sims, Francis). The number of infants who had neurodevelopmental follow-up is limited. It is possible that certain infants at greatest risk of adverse neurodevelopment were not resuscitated in some centres.(Finer, Jankov) Standardisation, timing of ultrasound and neurodevelopmental examination limit external validity and comparison of results. Rates of IVH and neurodevelopmental outcomes can also appear more or less favourable depending on whether the denominator used for calculation is CPR recipients or CPR survivors.(Jankov)
It is difficult to evaluate the effectiveness of initial airway resuscitation and the quality of CPR for infants described. It is probable that not all infants required CPR due to several having higher than expected Apgar scores.(Finer, Jankov, Sánchez-Torres). The incidence of CPR administration is high in certain centres.(Sánchez-Torres , Francis) Available evidence is inadequate to separate the contribution to adverse outcomes of the CPR itself and the factors which necessitated it. )
Current evidence is insufficient to accurately assess the effects of CPR at delivery on the outcome of extremely preterm infants. In practice, infants born at less than 26 weeks gestation are offered CPR and/or epinephrine at birth in England. In the 2006 EPICure cohort, CPR and/or epinephrine at birth was administered to 11% of actively supported infants at 23 weeks, 15% at 24 weeks, 11% at 25 weeks and 6% at 26 weeks.18 Prospective data for these CPR recipients and other extremely preterm infants receiving CPR at birth (after well performed airway and ventilation resuscitation) would be valuable in order to clarify outcomes.
Editor Comment
CC chest compressions; US ultrasound; IVH intraventricular haedmorrhage; PVL perventricular leucomalacia; HPI haemorrhagic periventricular infarction; IQR interquartile rangeClinical Bottom Line
Improvements in perinatal care have resulted in very preterm infants who received cardiopulmonary resuscitation (CPR) at birth surviving. (Grade C)There is limited retrospective evidence regarding the effect of CPR at birth on the survival and neurodevelopmental outcome of extremely preterm infants.
Small retrospective studies have reported normal neurodevelopment in one to two fifths of very preterm infants who received CPR at birth. (Grade C)
References
- Davis DJ How aggressive should delivery room cardiopulmonary resuscitation be for extremely low birth weight infants? Pediatrics 1999;104(3 Pt 1):428-34
- Finer NN, Horbar JD, Carpenter JH Cardiopulmonary resuscitation in the very low birth weight infant: the Vermont Oxford Network experience Pediatrics 1999;104(3 Pt 1):428-34
- Finer NN, Tarin T, Vaucher YE, et al Intact survival in extremely low birth weight infants after delivery room resuscitation Pediatrics 1999;104(4):e40
- Jankov RP, Asztalos EV, Skidmore MB Favourable neurological outcomes following delivery room cardiopulmonary resuscitation of infants < or = 750 g at birth. J Paediatr Child Health 2000;36(1):19-22
- Sánchez-Torres AM, García-Alix A, Cabañas F, et al [Impact of cardiopulmonary resuscitation on extremely low birth weight infants] An Pediatr (Barc) 2007;66(1):38-44
- O’ Donnell A, Gray P, Rogers Y. Mortality and neurodevelopmental outcome for infants receiving adrenaline in neonatal resuscitation. .J Paediatr Child Health 1998; 34: 551-556.
- Sims DG, Heal CA, Bartle SM. Use of adrenaline and atropine in neonatal resuscitation. Arch Dis Child 1994;70:F3–F10.
- Francis P, Casiro O. Neurodevelopmental outcome of very low birth weight children requiring chest compressions immediately after birth Pediatr Res 1997; 41:196.
- Jain L, Ferre C, Vidyasagar D, et al. Cardiopulmonary resuscitation of apparently stillborn infants: survival and long-term outcome J Pediatr 1991;118:778–82.
- World Health Organisation. ICD-10: disorders related to length of gestation and fetal growth. WHO http://apps.who.int/classifications/apps/icd/icd10online/ (Accessed 2 April 2010).
- Nuffield Council on Bioethics. Critical care decisions in fetal and neonatal medicine: ethical issues. http://nuffieldbioethics.org (Accessed 30 March 2010). 2006.
- Wilkinson AR, Ahluwalia J, Cole A, et al. Management of babies born extremely preterm at less than 26 weeks of gestation: a framework for clinical practice at the time of birth. Arch Dis Child Fetal Neonatal Ed 2009;94:F2–5.
- Ziino AJ, Davies MW, Davis PG. Epinephrine for the resuscitation of apparently stillborn or extremely bradycardic newborn infants. Cochrane Database Syst Rev . 2002;3:CD003849
- Perlman JM, Risser R. Cardiopulmonary resuscitation in the delivery room. Associated clinical events. Arch Pediatr Adolesc Med 1995;149:20–5.
- International Liason Committee on Resuscitation. International consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Part 7: Neonatal Resuscitation. 2005;67:293–303
- Costeloe K, Draper ES, Myles J, et al. Interventions to stabilise extremely preterm infants at birth. Arch Dis Child Fetal Neonatal Ed 2008;93:Fa9.