Sunday, March 18, 2007

Pine fresh, Meadow thick

As one of my colleague's has noted, Sally Clark—who was wrongly convicted of murdering two of her children—has died at home.
Mrs Clark served more than three years in jail after she was found guilty of smothering eight-week-old Harry and 11-week-old Christopher at her trial at Chester Crown Court in 1999.

The statement continued: "Sally was released in 2003 having been wrongfully imprisoned for more than three years, falsely accused of the murder of her two sons.

"Sadly, she never fully recovered from the effects of this appalling miscarriage of justice."

On appeal, Mrs Clark was found to have been wrongly convicted of the twin murders after new medical evidence emerged that had not been presented at her trial.

The fucking idiot who was greatly responsible for her conviction (although a woeful lack of understanding of statistics also played its part) was Professor Roy Meadow.

Professor Meadow was also involved as an expert witness in many other trials and, after the Clark case, other convictions were quashed.
He gave expert evidence in the trials of Sally Clark, Angela Cannings and Donna Anthony.

All three women were convicted of killing their children - and all three have subsequently been exonerated by the Court of Appeal after lengthy periods in prison.

These women were convicted almost solely on the expert testimony of Professor Meadow whostated, in the Clark case, that:
At her trial, Sir Roy said the odds of two children from such an affluent family dying of natural causes were one in 73 million.

This was subsequently perceived to be a huge pile of stinking horseshit. Quite apart from the fact that—as the Royal Statistical Society wrote in a letter of complaint to the Lord Chancellor—this figure had "no statistical basis", dear Roy forgot to tell us all what the statistical chances of a woman killing two of her children were.

But, you see, anyone can make this kind of mistake (although they fucking well shouldn't). The reason that I rage at Roy Meadow is his utter lack of scientific rigour.
Meadow's calculation had assumed that the cot death probability within any single family was the same as the aggregate ratio of cot deaths to births for the entire affluent-non-smoking population. No account had been taken of conditions specific to individual families (such as the hypothesised cot death gene) which might make some more vulnerable than others. The occurrence of one cot-death makes it likely that such conditions exist within the family in question, and the probability of subsequent deaths is therefore greater than the group average (estimates are mostly in the region of 1:100).

As a doctor, Meadow should at least have considered a cot death gene; he certainly should have considered environmental factors. But he stubbornly stuck to his flawed figure and ruined the lives of at least three families before he was rumbled.

And what happened to him? After the ruining a good number of lives with his ignorance, stupidity and intransigence he was accused of misconduct and struck off. This decision was overturned on Appeal to the High Court. So, after destroying three families, Roy Meadow has lost precisely nothing (other than any future expert witness fees, obviously).

So, it makes my blood boil when I see crap like this being written.
Perhaps [Clark] was possessed by guilt that she really had killed her kids and remorse that she had brought down two eminent professors of paediatrics in getting the decision reversed.

Really. Or maybe, just maybe, after having lost her babies and spent three years in jail for a crime that she did not commit, she was just not terribly well.

As for bringing down "two eminent professors of paediatrics"? Don't make me fucking laugh! Roy Meadow, a man so stupid and convinced of his rectitude that he refused to consider any genetic component to cot death? A man who proposed a theory (Munchausen Syndrome by Proxy) and then apparently destroyed the research?

What a very strange definition of "eminent" we have these days. If Sally Clark could be proud of anything now, the fact that she destroyed the credibility (and, hopefully, career) of this desperate charlatan should be one of her crowning glories.

UPDATE: sheer eloquence from The Englishman. Go read...

7 comments:

Shutter said...

You would probably benfit from looking at the following - take them in reverse or historical sequence.

You may find that your views on the conduct of the Trial and Appeals of Sally Cark miggt be at variance with the Truth.

JuliaM said...

Looks like that blogger had second thoughts about the post...that page is now 404'ing.

CoralPoetry said...
This comment has been removed by the author.
DrD said...

Your blood may be boiling but the rise in temperature has cooked the organ needed for rational analysis of facts. It's difficult to see how much further from the truth you could possibly stray, whether from consumate ignorance or willful mendacity. I suspect the former.There aren't any 'cot death genes' for if there were it wouldn't be a cot death. The figures considered many environmental factors: age of mother, income. smoking etc. They were not 'his' figures but produced by a statistician working for the government sponsored survey of sudden death in infancy. The chance of having a baby that dies in its cot in the night with no evidence of disease is about 1 in 8500(observed)The chance of that tragedy happening again in an affluent middle class household with a mother over 25 and a non smoker is remote.If you are dimwitted, fond of dope and a succession of thugish boyfriends then your chances could be as high as 1 in 200 but if you are at the other end of the scale then you could be back to 1 in 8500.Sally Clark was not convicted because of this minor piece of evidence but because her two boys were damaged, in Harry's case seriously so. Sally Clark's second appeal was allowed because the Court was told that path' results hadn't been made available at the trial. The lawyers dumped the stats and moved rapidly, as lawyers do, to make as much out of this as possible. The truth is immaterial.They were able to suggest that a baby boy who died in the space of a few minutes whilst his father went to get more food had succumbed to staphylococcal septicaemia, a diagnosis clinically absurd except to the medically illiterate, the ignorant and BBC journalists.Meadow's evidence had nothing to do with the allowing of Angela Canning's appeal and the allowing of an an appeal does not carry with it a presumption of innocence.
Munchausen's Syndrome by Proxy is described all over the world and many babies are alive because of the reluctant conclusion of one of our best paediaticians( they look after sick babies) that the mother who should be his ally in the care of the distressed child could be the agent of its suffering. A man who has given his life to one of the most demanding disciplines of medicine is hounded and vilified by the worst of men and women whose gross self indulgent narcissism condemns babies to a terrifying death at the hands of those they trust the most.

Anonymous said...

DR D.

Both children died within 5 to 8 hours of having an outdated, mercury laden, supercharged, toxic level of inoculation.

The bruising/contusions were caused by resuscitation efforts by ALL concerned and has been proved beyond doubt.

If you were British you would know the meaning of a wassock. Look in the mirror. You wouldn’t be allowed in Britain (you\'ve tried) because you are a stalking pervert.

If you know so much about the case, because you say you are an eminent doctor, then why have you omitted the fact that there was a massive bruise on the baby\'s leg, which was caused by the first paramedic on the scene? The paramedic held the baby by the leg and turned him upside down, presumably to clear an assumed obstruction in the lungs. It is the fundamental ABC of resuscitation. First clear the Airways. Check Breathing. Apply CPR (external breathing help) once the airway is clear. Nobody could tell if there actually WAS congestion in the lungs but this method of resuscitation was applied anyway.

Anonymous said...

Dr. D

Seems ironic that David Southall, the professor who accused Sally
Clark’s husband of murdering his child whilst he was away from
home, has been involved with smothering experimentation WITHOUT
PARENTAL CONSENT.


http://edmi.parliament.uk/EDMi/EDMDetails.aspx?EDMID=31383&SESSION=875

EDM 2767

WORK OF PROFESSOR DAVID SOUTHALL
17.10.2006


That this House notes that according to the report written by
Professor David Hull for North Staffordshire Trust about the work
of Professor David Southall in the report written for the
University Hospital of North Staffordshire by Professor McLeish and
Dr Durbin, Professor McLeish said that Professor Southall `pursued
multiple clinical research studies that were poorly designed and
therefore were unlikely to produce new knowledge of worth. More
worryingly he appears to have had insufficient regard for the
ethical standards that should surround all clinical studies in
babies'; believes that such comments are important comments that require proper consideration; is surprised that the University
Hospital of North Staffordshire is unable to find a copy of this
report; calls for the hospital to find a copy of this report and
publish its contents; and further calls for an independent judicial
or Parliamentary inquiry into the research and clinical activities
of Professor David Southall, the failure of the regulatory system
to prevent unethical experiments on babies managed by Professor
Southall and the misuse of child protection and judicial procedures
both to prevent parents from raising complaints about his research and procure children for his research.

http://www.guardian.co.uk/uk_news/story/0,,1859611,00.html

Monday August 28, 2006
Guardian

Detectives have stepped up an investigation into claims that the
leading paediatrician David Southall left a child brain damaged as a result of a controversial breathing experiment 15 years ago, the Guardian has learned.

South Wales police have broadened their inquiry into an allegation
that Professor Southall assaulted the boy by carrying out the test
and are asking dozens of parents whose children may have come into
contact with the paediatrician over the years to come forward if
their child suffered any injuries as a result of his treatment.
Professor Southall has denied that his treatment has harmed any
child.

In a letter to parents last week, Detective Inspector Chris
Mullane, of the force's child protection unit, said further
inquiries could be opened as a result of the responses from
parents. The letter says police are investigating an allegation of
assault on a boy that may have occurred while he was undergoing
treatment by Prof Southall at the University Hospital of Wales. It
asks parents: "Has your child been treated directly or indirectly
by Professor Southall ... Did your child suffer any injuries or
adverse effects from that treatment ... Have you reported this matter to the police or any other body?"

The investigation began after the parents of Ben McLean alleged
that he had been left brain damaged by Prof Southall's experiments at the University Hospital of Wales in 1991.

The child's mother, Davina McLean, believes that without their
informed consent, her five-year-old son was given carbon dioxide to breathe and his airway was occluded during a sleep study. She
claims that she and her husband were forced to take part in the
study after Prof Southall said they were suffering from
Munchausen's syndrome by proxy, and warned that unless they allowed Ben to take part he would be taken into care. Prof Southall has also denied these claims.

When Ben left hospital he was placed in foster care, but a year
later a court found the McLeans had not harmed their child. Ben,
now 20, lives with his parents and has severe speech and learning
difficulties. Mrs McLean told the Guardian: "We are pleased that
other parents out there who may have concerns are being contacted.
All we want is justice for our son."

Prof Southall has attracted praise and controversy during his long
career. Last year he was found guilty of serious professional
misconduct and banned from child protection work for three years
after wrongly accusing the husband of Sally Clark of killing their
baby sons.

http://www.bbc.co.uk/england/realmedia/midlandstoday/bulletin.ram

South Wales Police Heddlu De Cymru

Working with the Community Cydweithio Gyda’r Gymuned

Public Protection Unit
Central Police Station
King Edward VII Avenue
Cathays Park
CARDIFF
CF10 3NN

Telephone (029) 20527272

10th August 2006

Dear

South Wales Police are currently investigating an allegation of
assault on a young boy that may have occurred whilst undergoing
treatment by Professor David Southall at the University Hospital of Wales.

I have been given your details by Mr William BACHE, Solicitor, who
assures me that he has your authority for me to make contact with you.

I would be obliged if I could be provided with certain replies to
the below questions. I must emphasise that South Wales Police are not carrying out an enquiry into Professor Southall, but are
investigating one allegation of assault carried out in our force
area. It may well transpire that further enquiries are carried out
in the future if document dictate that to be the appropriate course
of action. Please reply via email if you wish or I have enclosed a
S.A.E.for your convenience.

1. Has your child been treated directly or indirectly by Professor Southall.
2. If yes please outline the document of that treatment.
3. Did your child suffer any injuries or adverse effects from that treatment.
4. Have you reported this matter to the Police or any other body
such as the GMC (please specify).
5. If you reported the matter to the police

i) which force
ii) when
iii) have you details of an investigating officer or any other
means of reference
iv) Result of the Police investigation

My apologies for being brief and to the point, but I am sure you
appreciate the complexities of this enquiry.

Yours faithfully


Chris Mullane
DETECTIVE INSPECTOR
chris.mullane@south-wales.pnn.police.uk



DAILY EXPRESS TUESDAY FEBRUARY 6 2001 Page 30

Scandal of 'smothered' babies in cot death test

Police investigate experiments on little children with lung problems

EXCLUSIVE
BY LUCY JOHNSTON
AND JONATHAN CALVERT


SECRET hospital cot death experiments in which doctors planned deliberately
to `smother' babies are being investigated by police.


The research project, devised by some of Britain's leading child
specialists, envisaged using tiny infants with severe breathing difficulties.

The babies' faces were to be covered with a mask attached to a breathing machine and their mouths `smothered' for up to 10 seconds on five occasions.


It is not clear whether the scheme was ever fully carried out, but
it appears that some parts did take place.

The controversial procedure, approved by an ethics committee, was regarded as safe. The infants would be secretly monitored by
doctors as they got older. If they died of unrelated illness,
pieces of their lungs, brains, livers, and hearts would be sent to a pathologist in Sheffield Children's Hospital for analysis and comparison with the project data.

The study was designed to help discover whether cot death was
caused by breathing and heart abnormalities and involved children across the country.

In a highly unusual move, doctors decided they would not seek
written consent from parents because they did not want to cause
alarm.

The study, named the Sudden Infant Death Project, was planned to be
carried out at three hospitals: Rotherham District General, the
Doncaster Royal Infirmary and the Barnsley District Hospital during
the late Eighties and early Nineties.

A spokesperson for the Rotherham District General Hospital said:
"Our consultant has said that the study did go ahead so I'm pretty
sure it did."

A spokeswoman for the Doncaster Royal Infirmary said it could not
comment on the matter "because it is subject of a police inquiry".

The Daily Express has evidence the experiment could also have been
conducted at other hospitals. Two sets of parents believe their
children were brain damaged after being put into similar
experiments.

The two children, whose brains were developing normally, now have
speech and co-ordination problems and severe learning difficulties.

The parents have not been able to find out what happened while
their children were in hospital.


British Medical Journal paper recording the experimentation in smothering. What higher authority to confirm this evil practice?

http://www.bmj.com/cgi/content/full/316/7135/887

http://tinyurl.com/2so6pd

Anonymous said...

Just in case the above British Medical Journal Item also disappears from the Internet here is an extract of the relevant parts:

BMJ 1998;316:887-894 ( 21 March )

Papers
Effect of exposure to 15% oxygen on breathing patterns and oxygen saturation in infants: interventional study
Commentary: Safety of participants in non-therapeutic research must be ensured
Commentary: Ethical approval of study was warranted
Authors' reply
Effect of exposure to 15% oxygen on breathing patterns and oxygen saturation in infants: interventional study
Editorial by Milner
K J Parkins, research fellow, a C F Poets, lecturer in paediatrics, b L M O'Brien, research assistant, a V A Stebbens, research assistant, a D P Southall, professor of paediatrics. a
a Academic Department of Paediatrics, North Staffordshire Hospital Centre, Stoke on Trent ST4 6QG, b Department of Paediatrics, Medical School, 30623 Hanover, Germany
Correspondence to: Professor Southall cai.uk@compuserve.com


Abstract
Objective: To assess the response of healthy infants to airway hypoxia (15% oxygen in nitrogen).
Design: Interventional study.
Settings: Infants' homes and paediatric ward.
Subjects: 34 healthy infants (20 boys) born at term; mean age at study 3.1 months. 13 of the infants had siblings whose deaths had been ascribed to the sudden infant death syndrome.
Intervention: Respiratory variables were measured in room air (pre-challenge), while infants were exposed to 15% oxygen (challenge), and after infants were returned to room air (post-challenge).
Main outcome measures: Baseline oxygen saturation as measured by pulse oximetry, frequency of isolated and periodic apnoea, and frequency of desaturation (oxygen saturation 80% for 4 s). Exposure to 15% oxygen was terminated if oxygen saturation fell to 80% for 1 min.
Results: Mean duration of exposure to 15% oxygen was 6.3 (SD 2.9) hours. Baseline oxygen saturation fell from a median of 97.6% (range 94.0% to 100%) in room air to 92.8% (84.7% to 100%) in 15% oxygen. There was no correlation between baseline oxygen saturation in room air and the extent of the fall in baseline oxygen saturation on exposure to 15% oxygen. During exposure to 15% oxygen there was a reduction in the proportion of time spent in regular breathing pattern and a 3.5-fold increase in the proportion of time spent in periodic apnoea (P<0.001). There was an increase in the frequency of desaturation from 0 episodes per hour (range 0 to 0.2) to 0.4 episodes per hour (0 to 35) (P<0.001). In 4 infants exposure to hypoxic conditions was ended early because of prolonged and severe falls in oxygen saturation.
Conclusions: A proportion of infants had episodes of prolonged ( 80% for 1 min) or recurrent shorter ( 80% for 4 s) desaturation, or both, when exposed to airway hypoxia. The quality and quantity of this response was unpredictable. These findings may explain why some infants with airway hypoxia caused by respiratory infection develop more severe hypoxaemia than others. Exposure to airway hypoxia similar to that experienced during air travel or on holiday at high altitude may be harmful to some infants.


Key messages

· A reduction in inspired oxygen concentration to 15% can induce severe prolonged hypoxaemia in a small proportion of infants
· Prediction of which infants will become hypoxaemic does not appear possible from analysing oxygenation or the respiratory pattern of infants breathing room air at sea level
· The way in which an infant responds to airway hypoxia may contribute to understanding the relation between respiratory infections, hypoxaemic episodes, and the sudden infant death syndrome
· Airline travel and holidays at high altitude may result in hypoxaemia in a small proportion of infants

Introduction

A reduction in the partial pressure of inspired oxygen may increase the risk of apparent life threatening events and sudden death in infancy.1-4 Airway hypoxia can be caused by respiratory tract infection.5 It may also be caused by a change to a higher altitude3 and air travel. The partial pressure of inspired oxygen on commercial aeroplanes is only 110 to 130 mm Hg; this corresponds to a fraction of inspired oxygen of 0.15 to 0.17 at sea level.6 Little is known about the physiological effects of airway hypoxia on respiratory function in infants. In adults acute airway hypoxia has pronounced effects on the control of breathing during sleep,7 and respiratory control and oxygenation are considered to be more vulnerable to the effects of hypoxia and other insults during infancy. We became interested in the effects of airway hypoxia on respiratory control in infants after two sets of parents attending our outpatient clinic reported that their infants had died of the sudden infant death syndrome after intercontinental flights; one infant had died between 14 and 19 hours after a flight and the other had died between 40 and 41 hours later.
In this study we exposed clinically healthy infants to 15% oxygen in nitrogen to discover the effects of airway hypoxia on arterial oxygenation and on the frequency of isolated and periodic apnoeic pauses. We also wanted to determine if there was a subgroup of infants who would develop potentially significant hypoxaemia during exposure to 15% oxygen.
Subjects and methods
Subjects
Thirty four infants (20 boys) were enrolled in the study. Twenty one were recruited by structured sampling of births at an obstetric unit run by general practitioners and 13 by approaching families who were receiving support in caring for an infant after a previous infant had died of the sudden infant death syndrome. The two groups were matched for age at the time of the study (mean age 3.1 months, SD 1.7 months for the group recruited from the obstetric unit and 1.8 months for the group of infants whose siblings had died of the sudden infant death syndrome). To be enrolled, infants had to have been born at term and have no history of respiratory distress or congenital anomalies; later, one infant was found to have ß thalassaemia minor but it was considered inappropriate to exclude him retrospectively. Twelve mothers had smoked during their pregnancies, half of these were mothers of children whose deaths had been ascribed to the sudden infant death syndrome.
In the week before the study no infant had had an illness with fever, but four developed respiratory infections; two additional infants had coughs from previous infections. One infant died suddenly three weeks after the study at age 2 months. Her two older half-siblings had allegedly died of the sudden infant death syndrome. All three deaths were later identified as infanticide.
We had intended to study infants who had undergone apparent life threatening events. The first four infants enrolled after such events, however, had abnormally low baseline values of oxygen saturation in room air and thus could not be subjected to airway hypoxia. Apparent life threatening events in two other infants were found to be due to epilepsy8 and intentional suffocation.9 For these reasons we decided to concentrate on infants without a history of apparent life threatening events.

Informed consent
Parents were sent a standard letter which briefly discussed the methods and purpose of the study, including the potential relevance of the research to the mechanism that might be responsible for some deaths from the sudden infant death syndrome. A self addressed envelope and reply form were included. If families were interested in participating they were contacted and arrangements were made to discuss the project in more detail. This happened either at the family's home or by telephone, and when possible both parents were involved. Information was presented to parents on the relation between the administration of 15% oxygen and airline flights, holidays at altitude, and the sudden infant death syndrome.
All parents were aware that an overnight physiological recording of their infant's oxygen saturation and respiratory variables would be done at home before their child was exposed to hypoxic conditions in hospital. Parents were informed that this recording would be analysed to ensure that values were within normal limits before the infant was exposed to 15% oxygen. All parents knew that an experienced paediatrician would be present throughout the infant's exposure to 15% oxygen, and that exposure would end if the infant's oxygen saturation dropped to 80% for 1 minute. Where applicable parents were informed that this had been necessary during previous recordings in this study. Parents were aware that they could withdraw their baby from the study at any time without explanation. After this discussion parents were given another information leaflet and were asked to sign a consent form. Each of the families in which exposure to 15% oxygen was ended early because of hypoxaemia of 80% for 1 minute was advised against taking their infants on flights or to high altitude until they were older than 12 months. This study was approved by the local research ethics committees.
Measurement of respiratory variables

Three tape recordings were made over two nights for each infant. Signals recorded were oxygen saturation in beat-to-beat mode (N200 pulse oximeter, Nellcor, Hayward, CA), pulse waveforms for validation of the accuracy of saturation measurements, and abdominal breathing movements with a volume expansion capsule placed on the abdominal wall (Graseby Medical, Watford). Recordings were made at 60 to 120 m above sea level. Infants were placed in their normal sleep position (lateral or supine). The first recording (pre-challenge) was made in room air in the infant's home; the results were checked to verify that the infant had normal baseline oxygen saturation values ( 94%) before the second recording. The second and third recordings were made in hospital 1 to 4 days later (median 26 h). The second recording (challenge) took place in an oxygen tent10 into which a medical gas mixture of 15% oxygen in nitrogen (British Oxygen Company, London) was delivered to maintain a monitored fraction of inspired oxygen of 0.15 to 0.16. Respired oxygen and carbon dioxide were monitored by a cannula on the upper lip (Elisa Duo, Engström, Stockholm) to confirm that rebreathing did not occur. Transcutaneous monitoring of the partial pressure of carbon dioxide was done at frequent intervals (Microgas, Kontron Instruments, Watford). Ambient temperature was maintained at 22°C to 26°C. Infants and monitors were observed continuously by an experienced paediatrician. According to our protocol, exposure to hypoxia would end if oxygen saturation fell to 80% for 1 minute. After the challenge infants were returned to room air and the third recording (post-challenge) was made throughout the rest of the night.
Recordings were printed and analysed manually by experienced technicians blind to the changes in inspired oxygen. Periods of regular and non-regular breathing patterns were identified11; a regular breathing pattern has been shown to be closely related to quiet sleep.12 Apnoeic pauses lasting 4 s were identified; these were classified by duration (4 s to 7.9 s, 8 s to 11.9 s, and 12 s13) and by whether they were isolated or appeared in periodic apnoea (episodes of three or more pauses, each separated by <20 breaths11).
Baseline oxygen saturation, heart rate, and respiratory rate were measured only during episodes of a regular breathing pattern.11 Periods when oxygen saturation fell to 80% for 4 s (desaturation) were identified throughout the recordings; these were classified as to whether they were associated with an apnoeic pause.13 Mean values of transcutaneous partial pressure of carbon dioxide were calculated.
Results are presented as median and range, or mean and standard deviation. Statistical analysis was performed using the Wilcoxon matched pairs test for paired data and the Mann-Whitney U test for the group comparisons. Correlations were assessed by Spearman's rank test.

Results
There was no significant difference in any variable between infants who were recruited from the obstetric unit and those from families in which an infant had previously died of the sudden infant death syndrome. Only two variables, respiratory rate and heart rate, were correlated with age. Results from the pre-challenge, challenge, and post-challenge recordings are shown in the table.


Results of tests on infants done before, during, and after exposure to 15% oxygen. Values are medians (range)


The mean duration of the pre-challenge recordings was 7.7 (SD 2.1) hours. Data from these recordings were similar to data already reported. 13 14 Baseline oxygen saturation was 94% in all infants, and only one infant had episodes of desaturation (three episodes, longest duration 11 s).
The mean duration of the recordings during the challenge was 6.3 (SD 2.9) hours. When compared with pre-challenge values, oxygen saturation during the challenge was lower (median difference 4.9%); this drop was highly variable (range 9.3% to 0.7%). Respiratory rates did not change significantly, but heart rates were 8 beats per minute higher (P<0.01); both rates were inversely correlated with age. Mean partial pressures of carbon dioxide during the challenge were within the normal range at 5.0 (SD 0.6) kPa. There was a significant decrease in the proportion of time spent in the regular breathing pattern, and a 3.5-fold increase overall in the proportion of time spent in periodic apnoea (P<0.001). There was a weak positive correlation between baseline oxygen saturation and amount of time spent in periodic apnoea (rs=0.44, P<0.01) during challenge. The frequency of isolated apnoeic pauses did not change significantly. Pauses tended to be shorter than during pre-challenge recording, with a decrease from 9.0% to 1.8% in the proportion lasting 8 s; none of the apnoeic pauses lasted 20 s.
There was a significant increase in the number of times desaturation occurred per hour during hypoxia (P<0.001); 21 out of 34 (62%) recordings had episodes of desaturation. A median of 96% of episodes of desaturation (range 16% to 100%) were associated with apnoeic pauses and were short (median average duration 5.0 s, range 4.0 s to 7.2 s). The median of the average of the lowest oxygen saturation value occurring during desaturation was 72% (67% to 76%).
The mean duration of the post-challenge recordings was 4.5 (SD 1.9) hours. All variables returned to pre-challenge values except for heart rate (which remained significantly raised) and the proportion of time spent in periodic apnoea (which was significantly reduced). Exposure to hypoxia was ended early in six infants. Analysis of the recordings showed that for four of the six the decision to end exposure early was appropriate. Oxygen saturation had been 80% for 1 minute in three infants. Oxygen saturation had been 80% for only 45 seconds in another infant but it had been <60% for two thirds of the time. Oxygen saturation values in the other two infants could not be interpreted because of movement artefact; a decision to withdraw these two infants from exposure to hypoxia was therefore inappropriate. However, while watching the monitoring the mother of one of these infants requested that her baby be returned to room air.
Withdrawal occurred after 1.9 to 5.2 hours (median 3.1 h) of hypoxic exposure in the four infants for whom it was appropriate; none of the infants woke spontaneously during their prolonged hypoxaemia. They were clinically well after withdrawal, although one received low flow oxygen (fraction of inspired oxygen 0.28) for the next hour to maintain oxygen saturation 94%. None had recently had a respiratory illness; one was the sibling of an infant who had died of the sudden infant death syndrome. Their ages were similar to those of the complete sample. Three of the four, however, had had low baseline values of oxygen saturation during the challenge; they were three of the six infants in the complete sample who had values <90% during the challenge. The fourth did not have any periods of a regular breathing pattern during the challenge so baseline values could not be measured. None of the four infants who were withdrawn from exposure had prolonged apnoeic pauses on their recordings.

Discussion
Main findings and limitations of the study

These healthy 1 to 6 month old infants had highly variable and unpredictable responses to acute airway hypoxia; some infants became hypoxaemic to such a degree that their exposure to hypoxia was ended.

Some limitations of this study should be considered. We gave priority to describing the effects of several hours of acute airway hypoxia on sleeping infants, rather than to identifying the mechanisms of the observed responses. We tried to interfere as little as possible with the infants' normal sleep patterns and decided against recording electroencephalograms, electro-oculograms, or quantifying ventilation. This prevented us from collecting precise information about the reasons why some infants developed severe hypoxaemia when exposed to 15% oxygen. Possible explanations include hypoventilation15 or increased inequalities between ventilation and perfusion.16 We do not know why the infants who became severely hypoxaemic did not wake up. We do not know whether our experimental conditions are identical to those of air travel and its effect on respiratory responses in infants. However, we could not find any data to suggest that there is a difference between a reduction in alveolar oxygen pressure due to reduced barometric pressure or due to a reduced fraction of inspired oxygen during constant atmospheric pressure. To address these issues infants would have to be studied during an airline flight or at high altitude.
Previous studies and possible relevance of these findings to the sudden infant death syndrome
Median values of baseline oxygen saturation during exposure to 15% oxygen in nitrogen in this study were similar to values measured by Lozano et al in 189 infants and young children born and living at 2640 m (93.3%, SD 2.1).17 The range of values found in the study of Lozano et al was much narrower than the range found in our study. This difference in interindividual variability in baseline values may have occurred because the infants studied by Lozano et al might have been both genetically and environmentally adapted to airway hypoxia, whereas our infants were not. This idea is supported by the results of a study done in Lhasa (altitude 3660 m) which found that indigenous Tibetan infants had mean oxygen saturation values of 87% to 88% during sleep, while Chinese infants, who had recently moved to the region, had values of only 76% to 80%.3 The lack of a genetic adaptation to high altitude has been proposed as the most likely cause for the disproportionately high rate of sudden deaths in infants soon after they have been moved to higher altitude. 3 4 High interindividual variability in the respiratory response to airway hypoxia may also explain why a proportion of infants with respiratory tract infections have low baseline values of oxygen saturation or an excessively high number of hypoxaemic episodes, or both.

There was no difference in the response to airway hypoxia in infants with a sibling whose death had been ascribed to the sudden infant death syndrome or in infants without such a family history. This is in accordance with other studies which failed to find evidence for a disturbance in respiratory control or function in the siblings of infants who had died of the sudden infant death syndrome, 18 19 and reinforces doubts about the appropriateness of using such infants for investigations into the pathophysiology of the syndrome.
The most frequent cause of airway hypoxia in infants is respiratory infection (particularly bronchiolitis). We and others have shown that a small proportion of infants with such infections may progress to developing life threatening hypoxaemic episodes. 5 20 Respiratory infections have also been linked with the sudden infant death syndrome in a number of studies.

Ethical issues

Was it ethically justified to expose healthy infants to 15% oxygen? Many infants travelling on aeroplanes or to holidays at high altitude are exposed to similar or even more markedly reduced partial pressures of inspired oxygen. Yet this exposure is considered safe. We were aware of anecdotal evidence of a small number of cases of the sudden infant death syndrome occurring after air travel, and of the observations made in Tibet.4 We considered that information on this important issue should ideally have been gathered before infants were permitted to travel by air. We found no evidence that such studies had been done. Information collected by British Airways showed that one infant had died during a flight from Hong Kong to Britain (NJ Byrne, personal communication). Our protocol was designed to allow us to identify immediately any potentially harmful degree of hypoxaemia, hypoventilation, or effects on cardiac rhythm; infants were observed continuously by an experienced paediatrician who followed strict guidelines on when to end an infant's exposure to hypoxia. We must also emphasise that although the siblings of infants whose deaths had been ascribed to the sudden infant death syndrome were already being monitored at home, the majority of the infants in this study had not been seen in our clinic before the study. Their families were, therefore, unlikely to feel conscious or unconscious pressure to comply with our request for participation.

Clinical implications

We have shown that a small number of infants may become hypoxaemic during several hours of exposure to a fraction of inspired oxygen of 0.15 to 0.16. We could not, for ethical and humanitarian reasons, determine whether this would have progressed to clinically apparent cyanotic episodes if exposure had continued. Unfortunately, there was no physiological or clinical variable in this study which would help identify infants who might develop clinically important hypoxaemia during later exposure to airway hypoxia. We believe that additional research is urgently needed into the effects on infants of prolonged airline flights or holidays at high altitude. Our findings may contribute to an understanding of the possible relation between respiratory infection with resulting airway hypoxia and some sudden deaths in infancy.