To Autoregulate or Not to Autoregulate—That is No Longer the Question

https://doi.org/10.1016/j.spen.2009.09.002Get rights and content

In the late 1970s, high cerebral blood flow was perceived as a cause of intracranial hemorrhage in the preterm infant. Intracranial hemorrhage was diagnosed by computed tomography and ultrasound found to be frequent not only in babies who died. Hemorrhage was soon linked to cerebral palsy in survivors. The analogy was hypertensive hemorrhagic stroke in the adult. Cerebral hemorrhage was perceived as the major (preventable) cause of brain injury in the preterm baby. An immature cerebral autoregulation or a vulnerability of the autoregulation exposed by preceding hypoxia or ischemia therefore became a focus of neonatal brain research in the 1980s. Over the years the focus has changed, first to the pathogenesis of hypoxic-ischemic brain injury, then to the effects of pCO2, and now 30 years later to a more comprehensive, less clearly hypothesis-driven exploration of the multitude of factors involved in cerebral blood flow and oxygenation. Meanwhile, some basic questions regarding autoregulation remain unanswered, and some concepts from the 1970s still direct clinical practice.

Section snippets

Survival of the Small Preterm with Brain Injury: The Neonatologist's Challenge

With rapidly evolving neonatal intensive care allowing survival of ever-smaller babies, the question of brain injury and survival of handicapped children became a prime concern for my mentor, Bent Friis-Hansen. He had been the head of the Department of Neonatology at Rigshospitalet since its establishment in 1965. Although his primary research field was body water during growth1 and nutrition, he had faced the practical aspects of neonatal care and soon supervised the implementation of

Cerebral Blood Flow: The Mecca was Copenhagen

Copenhagen in Denmark, 50 km south of the castle of Hamlet, was the origin of the use of radioactive isotopes to measure cerebral blood flow (CBF). Niels A Lassen4 obtained the first data in man in 1955 using the inert gas krypton. The inert gas xenon was first used to measure muscle blood flow by clearance after local injection. In 1965, his group published methods to measure regional cerebral blood flow in man by intracarotid injection5 or inhalation6 and by external detection of the gamma

A Simple Experiment

Thus, it was natural for Hans C. Lou, with an interest in pediatric neurology and newly employed in the Department of Neonatology, to take up the challenge to use the 133Xe clearance method to examine the role of CBF in perinatal brain injury. He placed an umbilical arterial catheter during clinical care to advance the catheter to the left carotid artery, the innominate artery or the aortic arch, and to inject 0.5 mCi 133Xe dissolved in 1 mL of normal saline. The catheter was withdrawn to its

The First Robust Data on CBF in the Human Newborn

The first publication from the collaboration of Lou et al7 was in 1977. It is the data from 8 newborns, shortly after birth. For Hans Lou, it was publication number 9, for Bent Friis-Hansen number 53, and for Niels Lassen number 238, available on PubMed. It thus came out of a significant collaborative experience in clinical science. The paper focuses on the surprisingly low values of CBF, in the newborn with values as low as 20 mL/100 g/min. The result of this “early slope” refers to the areas

The Change in Focus: From Ischemia to Hemorrhage

The next publication of the Copenhagen group included data from 19 newborns,10 of which 8 were the same as the previous publication. The correlation between CBF and blood pressure was now significant at the P < 0.0001 level, and the correlation was the same in infants with perinatal distress and those with postnatal distress only (presumed respiratory distress syndrome). The focus in this paper was on the impairment of autoregulation and on the risk of hemorrhage due to transmittance of

The Use of Sedatives to Limit Arterial Hypertension

The proposal of using sedatives to reduce the risk of hypertension was opposed as a potentially dangerous practice by Dr Wigglesworth13 in the correspondence section of the Lancet, stressing the danger of the Charybdis of hypotension and cerebral ischemia. The group from Copenhagen and Dr Wigglesworth's group from the Hammersmith in London had exchanges of letters of correspondence,14, 15 but agreed on the idea that the cerebral circulation was exquisitely sensitive to changes in arterial blood

My Haphazard Entry Into Neonatology

In February 1980, I was employed as a junior physician (reservelæge) in the Department of Neonatology. Unemployment among young medical graduates was high, and I got this opportunity partly because I had served 2 years as a district medical officer in Zambia, replacing my military service, and partly because I had shown pregraduate interest in research and published a paper on drug clearance as a measure of liver function; Friis-Hansen had worked in Zambia as a nutrition specialist for world

The Need for New Data—For a Noninvasive Method

At that time, Dr Wimberley was the research fellow of Dr Friis-Hansen Dr Wimberley's research area was the oxygen binding of hemoglobin F. As a side-job he took care of the new microcomputer-based continuous clinical data monitoring system of the department, and he was able to publish a weak correlation between arterial pressure peaks and cerebral hemorrhage.21 But what was really needed was more data on CBF. It was necessary to be able to measure CBF several times and at times other than

The Early Days of Computer-Based Signal Analysis

This was an opportunity. I became a research fellow in early 1982, I think because I had taken the ultrasound studies forward and because I had a bachelor's degree in computer science besides my medical degree and had begun helping out with the cerebrograph. It had come with a program implementing the inhalation/i.v. injection 133Xe clearance method for adults (Obrist). Although this method was used by Younkin et al22 from Philadelphia, and in a simplified version by Laura Ment,23 the analysis

An Improvised Cross-Atlantic Collaboration: Doppler Ultrasound vs Xenon Clearance

I presented some of my early methodological insights and solutions at the International Child Neurology Association meeting in Copenhagen in May 1982, to 10 people in a lecture theater. I realized that my data, the result of months of work, was not of interest to anybody. Over the years this impression has been reinforced; methodological scrutiny takes a quiet atmosphere. During the congress I contacted those 3 others presenting data on neonatal CBF: Henrietta Bada, Donald Younkin, and Patricia

An Afterthought on Methodology for Clinical Research

We continued to use Doppler ultrasound ourselves.26 I tried coanalysis of Doppler and blood pressure in the frequency domain, but in the end I became convinced that Doppler data on flow velocity could be impossible to interpret in CBF when arterial cross sectional area was dynamically regulated.27 Doppler was convincingly used to document hypoperfusion in the preterm with severe left-to-right shunt due to patent ductus28 and hyperperfusion in severe hypoxic-ischemic encephalopathy.29 These

A Valid “Non-Invasive” Measure of Cerebral Blood Flow in the Preterm Newborn

The brain–blood partition coefficient of xenon is less in the unmyelinated newborn brain compared with the adult, because xenon is lipophilic. I did the measurements by homogenizing fresh post-mortem specimens of white and gray matter. I settled for a method of analysis after more than a year of measurements, analyzing, programming, and simulation of sensitivity to errors. This method was derived from the two-compartment analysis, ie, it measures flow to the white as well as the gray matter of

Back to Cerebral Ischemia: Looking for the Evidence in the Form of Electrical Failure

During the mid-eighties, brain imaging and neurodevelopmental follow-up had moved the focus from IVH to periventricular leukomalacia; furthermore, the etiology of intraventricular intraparenchymal hemorrhage was no longer observed as a simple extension of the germinal layer hemorrhage,41 and hemorrhage without white matter damage was suggested to have no such importance for neurodevelopmental outcome. Thus, ischemic injury became the focus again, and I started to search for the electric

An Expensive Sidetrack: Flow Imaging

Niels Lassen was my official opponent, and during my thesis defense, just before Christmas 1989, he discussed the problem of assessment of blood flow to white matter, the tissue that seemed to be at real risk. As a result, Friis-Hansen and I obtained funding to have a single photon emission computed tomography scanner built particularly for the study of the newborn brain. The scanner was mobile and made it possible to come to the incubator and study ill infants without disturbing intensive

Confirming CO2 as the Major Determinant of CBF

Cerebral arteries and arterioles constrict with hypocapnia and dilate with hypercapnia. In normocapnic adults, small changes in arterial carbon dioxide tension (PaCO2) result in a change in CBF by 30%/kPa (4%/mm Hg). Similar reactivity in the normal term human neonate had been demonstrated by the method of venous occlusion plethysmography.48, 49 Although simple and quantitative, this method is dependent on the expansibility of the skull and has several limitations. Therefore, it was important

Hypocapnia, Periventricular Leukomalacia, and Cerebral Palsy

I realized that hypocapnia could be a most important risk factor for brain injury in the preterm infant. I had observed some cases of severe inadvertent hyperventilation when I looked for patients for my research projects. Thus, I planned a case-control study. I identified a group of 7 surviving infants discharged over a period of 18 months who had at least 1 arterial blood gas pressure with Pco2 less than 2 kPa (15 mm Hg) and 2 reference groups. Hans Lou did the neuropediatric evaluation,

A “Natural” Experiment

At Rigshospitalet, mild hyperventilation of preterm infants was intended during the early 1980s. The aim was to prevent IVH by extending the autoregulatory plateau.57 Furthermore, a new combined transcutaneous oxygen and CO2 electrode was used in the department. The electrode was developed by Radiometer, Copenhagen. The prototype instrument electrode showed the pCO2 on the screen without temperature correction, because it was still uncertain how the correction was to be performed. The pCO2 at

The CBF-arterial Blood Pressure Relation: To Autoregulate or Not

Neither my first results point to a direct association between arterial blood pressure and CBF35, 43 nor did those of Ole Pryds. Ole followed me as Friis-Hansen's research fellow, and in a large study on ventilated preterm infants he was able to discriminate between those who did not develop IVH (who did autoregulate) and those who did (who did not autoregulate).61 Furthermore, Ole studied asphyxiated term infants and found the same pattern; those who did not develop brain injury did

A More Useful Concept: Autoregulation as a Quantity Rather Than a Quality

In these studies, the presence of autoregulation was tested by regression of changes in CBF as a function of changes in blood pressure. This method was made possible by 2 or more measurements in each infant and by the fact that blood pressure changes spontaneously. A statistically insignificant regression coefficient was taken as absence of evidence for loss of autoregulation. More recent research uses near-infrared spectroscopy. This allows continuous monitoring of cerebral oxygenation, and

A Final Point of Relevance for Research Ethics and Clinical Practice

What matters is how much CBF can be increased by raising blood pressure in an ill preterm infant. Some data on this exists71, 72, 73, 74 and suggests that this increase is moderate. In this context, the paucity of experimental manipulation of blood pressure with CBF or cerebral oxygenation as the endpoint is a shame. Autoregulation has been tested by infusion of noradrenaline in adults with bacterial meningitis.75 In neonatal research, however, the idea from the 1970s of IVH caused by

Acknowledgment

Hans Lou, in particular, is thanked for his eager openness whenever I was critical about his earlier work.76 That has held true over the following years of collaboration. Hans has always worked for a new perspective and better insight.

References (76)

  • J.M. Perlman et al.

    The effect of patent ductus arteriosus on flow velocity in the anterior cerebral arteries: Ductal steal in the premature newborn infant

    J Pediatr

    (1981)
  • S. Ehehalt et al.

    Cerebral blood flow volume measurement with ultrasound: Interobserver reproducibility in preterm and term neonates

    Ultrasound Med Biol

    (2005)
  • O. Pryds et al.

    Preservation of single-flash visual evoked potentials at very low cerebral oxygen delivery in preterm infants

    Pediatr Neurol

    (1990)
  • G. Greisen et al.

    Discontinuous EEG activity, and periventricular brain injury in ill, preterm neonates

    Brain Dev

    (1989)
  • G. Greisen et al.

    Hypocarbia cause ischaemic brain damage in the preterm infant?

    Lancet

    (1986)
  • O. Pryds et al.

    Heterogeneity of cerebral vasoreactivity in preterm infants supported by mechanical ventilation

    J Pediatr

    (1989)
  • O. Pryds et al.

    Vasoparalysis is associated with brain damage in asphyxiated term infants

    J Pediatr

    (1990)
  • G. Greisen

    Autoregulation of cerebral blood flow in newborn babies

    Early Hum Dev

    (2005)
  • K. Lundstrøm et al.

    The haemodynamic effects of dopamine and volume expansion in sick preterm infants

    Early Hum Dev

    (2000)
  • B. Friis-Hansen

    Body water compartments in children: Changes during growth and related changes in body composition

    Pediatrics

    (1961)
  • R. Cooke et al.

    Respiratory failure in the newbornThe techniques and results of intermittent positive pressure ventilation

    Acta Pædiatrica Scand

    (1967)
  • J. Kamper et al.

    (Treatment of respiratory distress syndrome with continuous positive airway pressure)

    Ugeskr Laeger

    (1973)
  • N.A. Lassen et al.

    The cerebral blood flow in man determined by the use of radioactive krypton

    Acta Physiol Scand

    (1955)
  • K.B. Jensen et al.

    Regional cerebral blood flow determined by inhalation of xenon-133Summary of a methodological study

    Acta Neurol Scand Suppl

    (1965)
  • H.C. Lou et al.

    Low cerebral blood flow in hypotensive perinatal distress

    Acta Neurol Scand

    (1977)
  • G. Hambleton et al.

    Origin of intraventricular haemorrhage in the preterm infant

    Arch Dis Child

    (1976)
  • M. Osler et al.

    Prevention of the respiratory distress syndrome (RDS) by antepartum glucocorticoid (betamethasone) therapy combined with phenobarbitone and ritodrine

    Dan Med Bull

    (1978)
  • G. Greisen et al.

    Two-compartment analysis of plasma elimination of phenazone in normals and in patients with cirrhosis of the liver

    Acta Pharm Toxicol

    (1976)
  • G. Greisen et al.

    Intraventricular haemorrhage and method of delivery of VLBW infants

    J Perinat Med

    (1983)
  • P.D. Wimberley et al.

    Hypertensive peaks in the pathogenesis of intraventricular hemorrhage in the newbornAbolition by phenobarbitone sedation

    Acta Paediatr Scand

    (1982)
  • D.P. Younkin et al.

    Noninvasive method of estimating human newborn regional cerebral blood flow

    J Cereb Blood Flow Metab

    (1982)
  • L. Ment et al.

    Alterations in cerebral blood flow in preterm infants with intraventricular hemorrhage

    Pediatrics

    (1981)
  • G. Greisen et al.

    Analysis of cranial 133 xenon clearance in the newborn by the two-compartment model

    Scand J Clin Lab Invest

    (1984)
  • G. Greisen et al.

    Cerebral blood flow in the newborn infantComparison of Doppler ultrasound and 133 xenon clearance

    J Pediatr

    (1984)
  • G. Greisen et al.

    Poor reversibility of EEG abnormality in hypotensive, preterm neonates

    Acta Paediatr Scand

    (1988)
  • G. Greisen

    Analysis of cerebro-arterial Doppler flow velocity waveforms from newborn infants: Towards an index of cerebro-vascular resistance

    J Perinat Med

    (1986)
  • M.I. Levene et al.

    Comparison of two methods of predicting outcome in perinantal asphyxia

    Lancet

    (1985)
  • M.C. Toet et al.

    Cerebral oxygenation and electrical activity after birth asphyxia: Their relation to outcome

    Pediatrics

    (2006)
  • Cited by (0)

    View full text