The Obstetrician & Gynaecologist

10.1576/toag.13.3.169.27669 http://onlinetog.org

2011;13:169174

Review

Review How long have we got to get the baby out? A review of the effects of acute and profound intrapartum hypoxia and ischaemia
Authors Janet Rennie / Lewis Rosenbloom

Key content:
Intrapartum hypoxic ischaemia can damage the neonatal brain. There are two main models: acute profound and prolonged partial. Acute profound hypoxic ischaemia can damage the neonatal brain very quickly, with permanent brain injury occurring in some babies after >10 minutes of profound circulatory collapse. Magnetic resonance imaging is a powerful tool in the recognition of the pattern of acute profound hypoxic ischaemia.

Learning objectives:
To learn about the role of magnetic resonance imaging in diagnosing acute profound hypoxic ischaemia. To be aware of the literature on fetal hypoxic ischaemic brain damage in animal models and human babies.

Ethical issues:
How can we advance knowledge in this eld now that primate experiments are no longer considered ethical? Are there alternatives to animal experiments that would produce equally valid results? Keywords bradycardia / brain damage / cardiotocography / magnetic resonance imaging
Please cite this article as: Rennie J, Rosenbloom L. How long have we got to get the baby out? A review of the effects of acute and profound intrapartum hypoxia and ischaemia. The Obstetrician & Gynaecologist 2011;13:169174.

Author details
Janet Rennie MA MD FRCP FRCPCH DCH Consultant and Senior Lecturer in Neonatal Medicine Elizabeth Garrett Anderson Obstetric Hospital, University College London Hospitals, 2 North 250 Euston Rd, London NW1 2PQ, UK Email: [email protected] (corresponding author) Lewis Rosenbloom FRCP FRCPCH Honorary Consultant in Paediatric Neurology Royal Liverpool Childrens NHS Foundation Trust, Alder Hey Hospital, Eaton Road, West Derby, Liverpool, L12 2AP, UK

2011 Royal College of Obstetricians and Gynaecologists

169

Review

2011;13:169174

The Obstetrician & Gynaecologist

Introduction
There is no longer any doubt that some babies sustain permanent brain injury as a result of intrapartum hypoxia-ischaemia; the spectrum of end-stage disability that is recognised to be a consequence of this process has widened in recent years.1 Magnetic resonance imaging (MRI) of the brain has been proven to be a very powerful tool. Many babies with a perinatal history consistent with intrapartum hypoxia who have birth depression, early metabolic acidosis and encephalopathy, and whose MRI shows damage in the borderzone areas of the brain, are now considered to have acquired their damage as a result of hypoxic ischaemia even if their adverse outcome is limited to learning disability without any motor impairment.2 In this respect thinking has moved on considerably from the rigid criteria proposed by an international consensus group,3 whose essential criteria for intrapartum hypoxic insult specied that the adverse outcome was limited to spastic quadriplegic or athetoid cerebral palsy. The neonatal brain can be damaged very quickly indeed when the cause is acute near-total or acute

profound hypoxic ischaemia. Rapid recognition and response times in such potentially damaging situations remain extremely challenging for the obstetrician. An equally difficult problem is presented by a cardiotocograph (CTG) that is consistent with, but not diagnostic of, fetal hypoxic stress that may eventually be sufficient to cause fetal brain damage as a result of prolonged partial hypoxic ischaemia. The CTG is, of course, known to be an overly sensitive tool and a very poor indicator of fetal neurological status.4 With the benefit of hindsight, however, once the outcome is known the CTG is often the only tool with which to attempt a temporal reconstruction of any fetal decline. In this paper we review the literature that reects on the time to fetal hypoxic ischaemic brain damage in animal models and human babies. We have restricted our comments to the acute profound hypoxic ischaemic model.

Acute profound or acute neartotal brain injury

Evidence of the way that the mature human baby reacts to hypoxia combined with ischaemia has now accrued from relatively large numbers of MRIs from babies in the neonatal period and beyond. There are two basic patterns of damage seen as a result of intrapartum hypoxic ischaemia at term. The rst pattern is usually termed acute profound damage (Box 1 and Figure 1). The second pattern involves damage to the white matter in the borderzones between the vascular territories of the major cerebral arteries and is termed prolonged partial damage; we have not considered it further. Other patterns are rarely seen and are beyond the scope of our review.

Box 1

The pattern of acute profound damage caused by intrapartum hypoxic ischaemia at term

Damage occurs to the deep grey matter, the thalami and the perirolandic cortex; sometimes the hippocampi and cerebellar vermis are involved. Basal ganglia and thalamic lesions are regarded as the imaging signature of hypoxic ischaemic sentinel events. The clinical abnormality seen as a consequence is most commonly extrapyramidal motor dysfunction. Depending upon the degree and extent of the brain damage the least severely affected children may exhibit subtle features of a dyskinetic (athetoid) cerebral palsy with preservation of social and cognitive functioning. More severely affected individuals are likely to be profoundly disabled with immobility, a very high degree of dependence and commensurate care needs.

Figure 1

(a) Axial section of T2-weighted MRI scan showing abnormalities (white areas) in the posterior putamina and the ventrolateral nuclei (b) Diagram corresponding to the plane of section in Figure 1a, showing the damaged areas in heavy black

170

2011 Royal College of Obstetricians and Gynaecologists

The Obstetrician & Gynaecologist

2011;13:169174

Review

Animal models of acute profound damage

Animal models used to mimic acute total damage tend to involve hypoxic ischaemia sufcient to produce an oxygen saturation level of 30%, a 50% drop in heart rate and a virtual cessation in blood pressure. In monkeys the classic experiments on acute profound damage are those of Myers,9,10 in which asphyxia was produced by tying the umbilical cord completely and placing a rubber bag over the head of the fetus to prevent breathing; this insult produced bradycardia, gasping and metabolic acidosis within minutes (the pH fell from 7.3 to 6.8 in about 12 minutes). Ranck and Windle11 detached the entire placenta in near-term fetal monkeys and delivered them 1116 minutes later. In spite of complete cessation of the circulation the heart continued to exhibit an electrical discharge at 6070 beats per minute (bpm) and the circulation could be restored by resuscitation; when the baby monkeys were resuscitated they took >17 minutes to establish spontaneous breathing. Monkeys subjected to >2025 minutes of such an insult could be resuscitated but died later. Brain damage was seen after 10 minutes.10 Windle described his experiences with a similar experimental model,12 in which asphyxiation for >7 minutes invariably produced at least transient neurologic signs and permanent brain damage. The areas of the brain that were damaged after this catastrophic asphyxial insult included the basal ganglia and thalami together with the inferior colliculi: Myers10 reported a monotonously repetitive rank order, with the colliculi and posterior and lateral ventral thalamic nuclei damaged after 1013 minutes and more widespread thalamic involvement after 1618 minutes. Barbiturate anaesthesia extended the time to damage, as did hypothermia. Total asphyxial damage often also included the brain stem and on rare occasions the brain stem is also involved in human babies.13 These data are now almost 40 years old, many of the animals were anaesthetised and modern neonatal intensive care probably achieves survival of babies whose insult was equivalent to that which inevitably proved fatal in these early primate experiments. Nevertheless, these experiments have stood the test of time and the results remain extremely valuable. Windles careful descriptions11,12 of the evolving neurological syndrome he observed as the asphyxiated monkeys matured remain highly relevant today. Initially, the infant monkeys were helpless, excessively sleepy and had seizures, they took a long time to acquire the ability to drink milk from a bottle and they lacked curiosity; later they had an inability to grasp and reach, with athetoid movements. Eventually, 810 years later, the monkeys adjusted to their decits and appeared supercially normal but they
2011 Royal College of Obstetricians and Gynaecologists

still had difculties with tasks such as picking up raisins and could not retain simple information as well as their non-asphyxiated peers. More recently, results of umbilical cord clamping experiments using a primate model have been reported.14 Pregnant monkeys who were near term were anaesthetised with isourane and underwent surgery in which the umbilical cord was exteriorised and then clamped for 1215 minutes. The monkey fetuses were then delivered by caesarean section and resuscitated with positive pressure ventilation, undergoing electroencephalography and MRI in the neonatal period. After 15 minutes of clamping, the umbilical cord pH was 6.86 with a base decit of 23 mmol/l and the animals took between 1330 minutes to take their rst breath. The authors found that 12 minutes did not produce a reliably severe insult using this model, whereas 15 minutes did. These results conrm the very short time between no damage and damage: after 15 minutes of clamping 100% of the animals had signicant brain injury. Modern primate experiments cannot be conducted without anaesthesia and isourane binds to gamma-amino butyric acid (GABA), glutamate and glycine receptors; some studies suggest neuroprotection. Primate experiments are probably the most relevant, but many other animal models of neonatal hypoxic ischaemia have been developed using piglets, lambs, gerbils, mice and rats. Some argue that the piglet brain is closest in maturation to the human newborn.15 Cord occlusion for 10 minutes in fetal sheep resulting in acidosis (pH 6.9, lactate 6 mmol/l) and hypotension was followed by brain damage, predominantly in the hippocampus but also in the basal ganglia and thalamus.16 The blood pressure was not completely abolished, but was reduced to about 50% of the pre-occlusion level, and the heart rate fell from about 175 bpm to about 75 bpm in these lambs. Occlusion of the carotid artery of fetal sheep (after ligation of the anastomoses between the carotid and vertebral circulation) for 30 minutes produced an electrically silent electroencephalograph which persisted for 79 hours after release of the occlusion and was followed by delayed seizures and damage in the parasagittal cortex.17 Mild selective neuronal loss in the thalami and striatum was seen after only 10 minutes of occlusion in this experimental model (Figure 2). Repeated short umbilical occlusions can also cause damage to the striatum (equivalent to the deep grey matter) in lambs and may be closer to the insult that often affects the human fetus.16 In these experiments, four periods of 5 minutes of total umbilical cord occlusion produced neuronal damage in the striatum, with selective loss of GABAergic neurons in the basal ganglia. Selective
171

Review

2011;13:169174

The Obstetrician & Gynaecologist

Figure 2

Neuronal damage in nine brain regions following increasing durations of ischaemia, ranked in order of total damage scores on a linearised scale of 0100. The insult involved ination of carotid artery cuffs in fetal sheep for 0, 10, 20, 30 or 40 minutes.17 Reproduced with permission from the American Neurological Association

neuronal loss, predominantly in the hippocampus but also in the striatum and parasagittal cortex, was consistently seen after an isolated 10-minute episode of umbilical cord occlusion in near-term sheep fetuses (two of the nine did not recover from the insult).18 While all these animal data were obtained from experiments that are designed to provide a reproducible severe acute asphyxia, they may not always be directly referable to the nonanaesthetised human fetus subjected to a uterine rupture, cord prolapse, massive placental abruption, cord occlusion or shoulder dystocia. Nevertheless, the information obtained is valuable and there are many similarities to the human situation. The newborn animals developed a marked acute metabolic acidosis, required resuscitation, developed an early encephalopathy and acquired damage to the deep grey matter of the brain. Virtually all were permanently and seriously damaged after an insult lasting 1215 minutes and many were damaged after 10 minutes (some after 7) or repeated episodes of 5 minutes of umbilical cord occlusion. Animal experiments are costly and difcult to perform and the aim of a researcher in this eld is to produce a reliably severe lesion in virtually all animals, not to use an insult which leaves up to half the animals undamaged. In general, of course, in the human situation an intrapartum hypoxic ischaemic insult is rarely isolated in the way we have described and it is important to note that fetal monkeys who were
172

already acidotic because of a period of partial asphyxia acquired damage to the deep grey matter after a very short period of superimposed acute asphyxia, lasting 34 minutes,9 and some piglets cannot be resuscitated after short insults. The animal data we reviewed have been used to synthesise a 10-minute rule which reects that, on the balance of probability, a human fetus will acquire damage of the acute profound type after a short severe asphyxial insult lasting >10 minutes and that >50% of fetuses probably sustain some damage after an insult of this duration. We consider this to be a reasonable conclusion. It follows from this that the obstetrician probably only has a 10-minute period in which to deliver a fetus that has developed a sustained bradycardia. While the 10-minute rule has proved to be a useful rule of thumb, there is a degree of biological variability and variation in the severity of the insult and the prior state of the fetus. Hence it is helpful to re-examine the evidence from time to time, as we have done here. We have also re-evaluated the published human cases below.

Human cases with known time to damage in the acute profound model
Between us we, the authors, have studied hundreds of cases of children with damage to the deep grey matter acquired as a result of an acute profound hypoxic ischaemic insult which
2011 Royal College of Obstetricians and Gynaecologists

The Obstetrician & Gynaecologist

2011;13:169174

Review

occurred either during or before labour at term, but it is not always possible to time the damaging event with precision. Nevertheless, we have seen children with severe damage that was sustained after only 1015 minutes of bradycardia when the previous CTG was normal and others who appear to have survived, albeit often with very severe damage, after 4050 minutes. Given the frequency with which this pattern of damage is now recognised in clinical practice there are surprisingly few published cases in which the duration of insult has been addressed and more work is needed in this area in order to inform obstetric practice. Okumura and colleagues19 reported two cases in which fetal monitoring showed a fetal bradycardia for 20 minutes and in which there was MRI evidence of damage to the basal ganglia, thalami and around the central sulci. The rst was a boy whose CTG was entirely normal until he experienced a sudden fall in fetal heart rate to 60 bpm which lasted 30 minutes before recovering and who was delivered 36 minutes after the end of the bradycardia. Following his delivery by emergency caesarean section his Apgar scores were 2 and 4 at 1 and 5 minutes and he was hypotonic and jittery. He was tube fed, developed severe spastic quadriplegia and mental retardation and died at the age of 15 months. Naeye and Lin20 made a retrospective study of the notes of all children in their institution with cerebral palsy who were singletons born at >37 weeks of gestation and who had had a documented fetal bradycardia of <80 bpm which persisted until delivery (and usually fell to 60 bpm). Thirty-three of 55 such cases had damage in the deep grey matter and some children had additional lesions in the subcortical white matter. The shortest insult duration was 14 minutes, in a case of placental abruption. Children with longer insults tended to have damage both to the deep grey matter and to the subcortical white matter. Barkovichs group in San Francisco21 have impressive experience of MRI, studying over 6000 cases of neonatal encephalopathy in term and near-term babies. Their carefully performed and sequential MRI studies, which include spectroscopy, were targeted at determining when the insult occurred, not how long it lasted. These interesting serial studies show that some babies probably sustain their brain damage 23 days before birth but they do not shed much light on the duration of insult required to damage the neonatal brain. A review of the literature22 reporting fetal survival after maternal cardiac arrest reveals that few infants delivered >10 minutes after maternal collapse requiring resuscitation survive to be healthy individuals, although there were some normal survivors in the group delivered between 1115 minutes afterwards.
2011 Royal College of Obstetricians and Gynaecologists

Uterine rupture can be a catastrophic event for the fetus, particularly if it results in extrusion from the uterus; the human fetus is probably exposed to a situation that is close to that of the primate experiments. In a retrospective study23 of 106 cases of uterine rupture between 1983 and 1992, five babies were diagnosed as having asphyxia (there was no long-term follow up or MRI). A prolonged deceleration was defined as a fetal heart rate <90 for more than a minute, but there was only one such case in the asphyxiated group that did not have prior CTG abnormalities. The authors concluded that fetuses who were exposed to prolonged deceleration could survive without asphyxia if they were delivered before 17 minutes had elapsed, but several babies developed neonatal morbidity after 10 minutes of a prolonged deceleration if there had been previous late decelerations (range 3690 minutes). These data support the evidence provided by the primate studies regarding a worse outcome if an acute near-total insult is superimposed on a previous episode of partial hypoxia. More recently, a study of 23 cases of complete uterine rupture in Quebec24 showed that a diagnosis of hypoxic ischaemic encephalopathy was made in three and a further baby died. The three babies with hypoxic ischaemic encephalopathy were delivered 15,16 and 23 minutes after the onset of a prolonged fetal heart rate deceleration; all three were completely extruded from the uterus, had a metabolic acidosis at birth and developed motor delay. Phelan and Ahn25 found that permanent brain injury occurred after 18 minutes with a heart rate <60 bpm (this was the shortest known duration of insult in this study). Pasternak and Gorey5 describe a baby born after a uterine rupture and 1015 minutes of bradycardia who never developed the ability to feed and who died before discharge. We are aware that for obstetricians the accepted decision-to-delivery interval in the most urgent cases is 30 minutes and that the evidence provided in this paper is that this is too slow to avoid hypoxic brain damage in a small number of fetuses. From our perspective we recognise that there are problems in prospectively identifying the fetuses that fall into this very high-risk category. These problems include the limitations of current fetal monitoring and the practical difculties in achieving a more rapid decision-to-delivery interval in some units. Conversely, we appreciate that there is the potential for unnecessary intervention should an episode of fetal bradycardia cease spontaneously and that there is some risk to the mother of rapid general anaesthesia. Nevertheless, obstetricians need to be aware of the data detailed in this paper when considering how to ensure the best outcomes for both mothers and babies.
173

Review

2011;13:169174

The Obstetrician & Gynaecologist

Conclusion
Human cases of acute profound hypoxic ischaemia in which the insult duration can be timed with precision remain rare and there is often uncertainty about the prior state of fetal health. Nevertheless, it is the case that in many circumstances the human baby does appear to experience an insult that is close to that administered to animal fetuses in a laboratory setting. While there is clear variability both in the fetal reserve and in the duration and degree of the insult, we are now of the opinion that the concept that damage begins to accrue after 10 minutes of an acute profound hypoxic ischaemic insult, originally constructed from the results of the work of Windle and Myers, continues to serve well as a framework and, hence, all obstetricians need to be aware of these data. Fortunately, these cases remain rare and the only way forward is for a national register to be created so that experience can be shared and pooled for the benet of future generations. References
1 Rennie JM, Hagmann CF, Robertson NJ. Outcome after intrapartum hypoxic ischaemia at term. Semin Fetal Neonatal Med 2007;12:398407. doi:10.1016/j.siny.2007.07.006 2 Gonzalez FF, Miller SP. Does perinatal asphyxia impair cognitive function without cerebral palsy? Arch Dis Childhood Fetal Neonatal Ed 2006;91:4549. doi:10.1136/adc.2005.092445 3 MacLennan A. A template for dening a causal relation between acute intrapartum events and cerebral palsy: international consensus statement. BMJ 1999;319:10549. 4 Nelson KB, Dambrosia JM, Ting TY, Grether JK. Uncertain value of electronic fetal monitoring in predicting cerebral palsy N Engl J Med 1996;334:6138. doi:10.1056/NEJM199603073341001 5 Pasternak JF, Gorey MT. The syndrome of acute near-total intrauterine asphyxia in the term infant. Pediatr Neurol 1998;18:3918. doi:10.1016/S0887-8994(98)00002-2 6 Cowan F. Outcome after intrapartum asphyxia in term infants. Semin Neonatol 2000;5:127140. doi:10.1053/siny.2000.0011 7 Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, et al. Origin and timing of brain lesions in term infants with neonatal encephalopathy. Lancet 2003;361:73642. doi:10.1016/S01406736(03)12658-X

8 Okereafor A, Allsop J, Counsell SJ, Fitzpatrick J, Azzopardi D, Rutherford MA, et al. Patterns of brain injury in neonates exposed to perinatal sentinel events. Pediatrics 2008;121:90614. doi:10.1542/peds.2007-0770 9 Myers RE. Four patterns of perinatal brain damage and their occurrence in primates. Adv Neurol 1975;10:2234. 10 Myers RE. Two patterns of perinatal brain damage and their conditions of occurrence. Am J Obstet Gynecol 1972;112:24676. 11 Ranck JB, Windle WF. Brain damage in the monkey, macaca mulatta, by asphyxia neonatorum. Exp Neurol 1959;1:131150. 12 Windle WF. Brain damage at birth. JAMA 1968;206;1967. doi:10.1001/jama.206.9.1967 13 Natsume J, Watanabe K, Kuno, K, Hayakawa F, Hashizume Y. Clinical, neurophysiologic, and neuropathological features of an infant with brain damage of total asphyxia type (Myers). Pediatr Neurol 1995;13:6164. doi:10.1016/0887-8994(95)00054-J 14 Juul SE, Aylward E, Richards T, McPherson RJ, Kuratani J, BurbacherTM. Prenatal cord clamping in newborn macaca nemestrina: a model of perinatal asphyxia. Dev Neurosci 2007;29:31120. doi:10.1159/000105472 15 Thoresen M, Haaland K, Loberg EM, Whitelaw A, Apricena F, Hanko E, et al. A piglet survival model of posthypoxic encephalopathy. Pediatr Res 1996;40:73848. doi:10.1203/00006450-199611000-00014 16 Mallard EC, Waldvogel HJ, Williams CE, Faull RLM, Gluckman PD. Repeated asphyxia causes loss of striatal projection neurons in the fetal sheep brain. Neuroscience 1995;65:82736. doi:10.1016/03064522(94)00504-X 17 Williams CE, Gunn AJ, Mallard C, Gluckman PD. Outcome after ischemia in the developing sheep brain: an electroencephalographic and histological study. Ann Neurol 1992;31:1421. doi:10.1002/ana.410310104 18 Mallard EC, Williams CE, Johnston BM, Gluckman PD. Increased vulnerability to neuronal damage after umbilical cord occlusion in fetal sheep with advancing gestation. Am J Obstet Gynecol 1994;170:20614. 19 Okumura A, Hayakawa F, Kato T, Kuno K, Watanabe K. Bilateral basal ganglia-thalamic lesions subsequent to prolonged fetal bradycardia. Early Hum Dev 2000;58:111118. doi:10.1016/S0378-3782(00)00072-4 20 Naeye RL, Lin H-M. Determination of the timing of fetal brain damage from hypoxemia-ischaemia. Am J Obstet Gynecol 2001;184:217224. doi:10.1067/mob.2001.108996 21 Barkovich AJ, Miller SP, Bartha A, Newton N, Hamrick SEG, Mukherjee P, et al. MR imaging, MR spectroscopy, and diffusion tensor imaging of sequential studies in neonates with encephalopathy. Am J Neuroradiol 2006;27: 53347. 22 KatzVL, Dotters DJ, Droegemueller W. Perimortem cesarean delivery Obstet Gynecol 1986;68:5716. 23 Leung AS, Leung EK, Paul RH. Uterine rupture after previous cesarean delivery: maternal and fetal consequences. Am J Obstet Gynecol 1993;169:94550. 24 Bujold E, Gauthier RJ. Neonatal morbidity associated with uterine rupture: what are the risk factors? Am J Obstet Gynecol 2002;186:31114. doi:10.1067/mob.2002.119923 25 Phelan JP, Ahn MO. Fetal heart rate observations in 300 term braindamaged infants. J Maternal Fetal Invest 1998;8:15.

174

2011 Royal College of Obstetricians and Gynaecologists