I. BRAIN DAMAGE AT BIRTH

1 - Asphyxia at Birth
I first learned of the vulnerability of the auditory system to asphyxia at birth when the October 1969 issue of Scientific American arrived in my mailbox [1]. My son Conrad was then five years old and spoke only in parroted phrases. He loved music and sang all his favorite songs played on the radio. But his hearing was not normal. He became confused and upset in noisy environments and was terrified to enter a room where he saw a telephone. Trying to get him to listen to someone on the phone could also be upsetting.

Figure 1 (up top) is an autoradiogram from Kety's research (in cats) that shows the greatest amount of a radioactive tracer in the inferior colliculus one minute after injection – thus indicating greatest circulation in this small nucleus of the brainstem auditory pathway. Results of the first experiments on blood flow were confirmed using different radioactive tracers [5, 6]. Methods for glucose uptake and measurement of aerobic enzymes later revealed that aerobic metabolism is highest in the inferior colliculus and other brain areas of high circulatory rate [7-14]. Methods for both blood flow and glucose uptake have been widely used in research on neurochemistry [15-25]

Figure 2 (above) shows damage in the inferior colliculi of the brain of a monkey subjected to asphyxia at birth. The inferior colliculi are pitted by cavities (bilaterally) left by cells that disintegrated over the five year lifespan of the monkey. Figure 3 shows by way of contrast the appearance of the inferior colliculus in the brain of a normal monkey of the same age in which the densely packed neurons are intact.
[Top]

2 - Hypoxic Birth
In their initial experiments Windle and co-workers pulled a saline-filled rubber sac over the head of infant monkeys at birth and then clamped the umbilical cord [26, 27]. The lungs were thus prevented from taking in air, and blood flow from the placenta was abruptly blocked. The result was a sudden catastrophic cutoff of respiratory gas exchange (asphyxia). That this did not cause cerebral palsy came as a complete surprise!

Still looking for causes of cerebral palsy, Ronald Myers (a member of Windle's team) employed a technique to produce intermittent obstruction of umbilical blood flow late in gestation [2]. This somewhat less catastrophic disruption more closely mimics what happens during a difficult birth, in which a state of partial oxygen insufficiency develops (hypoxia). Even partial interference with oxygen delivery is not healthy, and cerebral palsy and the pattern of cortical damage long associated with cerebral palsy were observed in monkeys subjected to hypoxia in this way.

However, figure 4 is from Myers' 1972 paper, "Two patterns of perinatal brain damage and their conditions of occurrence," and confirms that damage to the inferior colliculus results when oxygen delivery is totally obstructed at birth.

Miller and Myers (1970, 1972) found the same pattern of brainstem damage also in adult monkeys subjected to cardiac arrest; and, as in the case of infant monkeys, found damage in the cerebral cortex in adult monkeys subjected to partial disruption of circulation [28, 29]. The infant heart withstands asphyxia longer, but the infant brain is no less vulnerable to damage than that of the adult [2]. Furthermore, the difficult to detect pattern of brainstem lesions in newborn monkeys resulted in a disruption of brain development; monkeys kept alive to maturity were found to have neuropathological changes in additional subcortical areas and the cerebral cortex [30].
[Top]

3 – Asphyxia Versus Hypoxia
Asphyxia is not just a more severe degree of hypoxia. Asphyxia is a different kind of insult. In situations of reduced blood flow or oxygen insufficiency (hypoxia), protective mechanisms go into effect. For example, respiratory gas exchange (oxygen for carbon dioxide) is a biochemical process that provides immediate adjustment during any hypoxic episode: Except in the lungs, hemoglobin releases oxygen in exchange for carbon dioxide; this mechanism (the Bohr effect) has been understood for nearly a century [31-36]. Organs of highest metabolic rate produce the greatest amount of carbon dioxide end-product; and these organs are first to obtain oxygen from hemoglobin, especially during any episode of oxygen (or circulatory) insufficiency.

The brain is a collection of separate sensory, motor, and association circuits with differing metabolic needs; nuclei within the auditory system have the greatest need for respiratory gas exchange. Thus during chronic compromise of circulation, hemoglobin will give up the oxygen needed by the inferior colliculus first, and then reach the so-called "watershed" areas of the cerebral cortex depleted of what oxygen it carried from the lungs or placenta. Cortical damage is the expected and most usual consequence of this kind of prolonged hypoxia [2].

Kusumoto et al. (1995) used an autoradiographic method to investigate cerebral blood flow in immature gerbils after bilateral occlusion of the carotid arteries [21]. In 2-week old gerbils 5 minutes of bilateral carotid occlusion produced severe forebrain ischemia. By way of contrast, blood flow in the inferior colliculus and nuclei of the pons increased dramatically (in the inferior colliculus from 91 to 124 ml per minute and in the pons from 61 to 90 ml).

Occlusion of the carotid arteries produced a severe but partial blockage of blood flow; otherwise measurement of residual circulation would not have been possible. The forebrain became severely ischemic at the expense of whatever protective mechanism went into effect to increase circulation to the inferior colliculus and nuclei of the pons. The effects of this change in circulation can be compared with the prolonged partial hypoxia Myers (1972) found caused damage to motor areas of the cortex in monkeys [2].

Asphyxia is catastrophic and most often fatal if it persists to the point of cardiac failure. Newborn monkeys could in most cases be resuscitated after up to 20 to 25 minutes of asphyxia [1, 2], but visible damage was evident in the brain if asphyxia lasted for eight to ten minutes. Later developing neuropathological changes were found throughout the brain in monkeys kept alive for several years, even in animals subjected to asphyxia of duration too short to produce visible brainstem lesions [30]. The commonly held assumption that infants withstand asphyxia and hypoxia better than mature individuals would appear to be a dangerous misconception, especially in view of the impact of early impairment on maturation of the brain.
[Top]

4 - Human Conditions
Myers rejected the idea that the brainstem pattern of damage had any relevance to what happens in human cases of hypoxic birth. But, figure 5 (below) shows damage of the inferior colliculi found in a human infant who died from accidental suffocation; and symmetric bilateral involvement of brainstem nuclei (including the inferior colliculi) has been noted in at least seven other reports of damage found following death in infancy related to injury at birth [37-43].

Saigal and Usher expressed the opinion that too large a transfusion of placental blood could put an infant at risk for polycythemia and jaundice [47]. Research by Windle and coworkers was apparently already part of forgotten history; Lucey et al. (1964) had demonstrated that even very high levels of bilirubin did not cross the blood brain barrier except in monkeys whose brains were already damaged by asphyxia [54].
[Top]

5 - Stages of Asphyxia
Myers believed that hypoxia during gestation was a more likely cause of brain impairment than asphyxia at birth, but he nevertheless thoroughly investigated what happens during asphyxia. His data can now be viewed as relevant to what must happen when the umbilical cord is clamped before a newborn infant's lungs are functional – and bring to mind the stages of drowning described by Sebastian Junger in The Perfect Storm [55].

Myers monitored heart rate, blood pressure, oxygen, carbon dioxide, pH, and respiratory efforts during asphyxia and resuscitation. He noted that envelopment of the fetal head and clamping of the umbilical cord led to an immediate increase in blood pressure followed within 20 seconds by a rapid decrease associated with a rapid fall in blood oxygen. Aerobic metabolism is completely halted by a minute and a half.

Myers commented that the early fall in blood oxygen is the most precipitous and dramatic of any of the changes during asphyxia and leads to deterioration of the heart and other organs of the body. Neurons then become dependent on glycogen stores within their surrounding astrocytes. Gasping began six minutes after cord clamping and continued to the fourteenth minute of asphyxia after which the animal entered a terminal state of apnea, becoming pulseless, pale, and flaccid. Windle noted minimal damage in two monkeys asphyxiated for only six minutes [1].

Visible lesions of brainstem nuclei were always seen in monkeys subjected to asphyxia of ten minutes duration in Myers' experiments. But microscopic examination revealed that damage to mitochondria and cell membranes preceded development of visible damage [2]. Damage of cell membranes is the basis for breakdown of the blood-brain barrier that protects against entry of substances like bilirubin from the blood stream. Compromise of brain and other organs clearly precedes the occurrence of visible damage.

Progressive long-term signs of damage throughout the brain were found by Faro and Windle (1969) in monkeys kept alive for several months or years following asphyxia [30]. Respiratory distress due to lung damage from asphyxia also led to more widespread damage. Damage from oxygen insufficiency is not ordinarily as "monotonously" predictable as found in the early experiments on asphyxia.

Respect for animal rights would prevent doing further experimentation of this kind with monkeys. But the data obtained deserves renewed attention. Myers demonstrated that prolonged partial disruption of circulation in utero leads to damage of the cerebral cortex and cerebral palsy, and he adamantly rejected the idea that brainstem damage caused by catastrophic asphyxia could be relevant to any human condition. But we are suddenly living in a time of increased developmental disorders, autism, childhood asthma, early gastrointestinal problems, and infant anemia. Immediate clamping of the umbilical cord at birth has become a standard procedure in the past 20 years and should be investigated as possibly contributory to rising rates of these childhood disorders.
[Top]

6 – The Umbilical Cord Lifeline
Clamping the umbilical cord at birth is a human invention. The umbilical cord is an infant's lifeline throughout gestation; it should go without saying that it remains the newborn's lifeline until lung function is established. Clamping the cord before a baby breathes can be expected to result in at least a brief period of oxygen deprivation. Seven to ten minutes of asphyxia resulted in visible damage in the brainstem in newborn monkeys, but Myers (1972) described microscopic changes following asphyxia of lesser duration. In human infants a low Apgar score at five minutes is ominous, and associated with autism [56, 57]. Even the briefest lapse of respiration should be avoided.

Looking back at historical textbooks on obstetrics, waiting at least for the infant to breathe on its own was traditionally always required before cutting the cord [58-65].

For example from 1850 to 1930:
"A strong healthy child, as soon as it is born, will begin to breathe freely, and in most cases cry vigorously. As soon as it has thus given satisfactory proof of its respiratory power, you may at once proceed to separate it from its mother by tying and dividing the umbilical cord." (Swayne 1856, p20)

"As soon as the child cries we may proceed to tie and separate the cord." (Playfair 1880, p283)

"The cord should not be tied until the child has breathed vigorously a few times. When there is no occasion for haste, it is safer to wait until the pulsations of the cord have ceased altogether."
(Lusk 1882, pp214)

"Immediately after its birth the child usually makes an inspiratory movement and then begins to cry. In such circumstances it should be placed between the patient's legs in such a manner to have the cord lax, and thus avoid traction upon it… Normally the cord should not be ligated until it has ceased to pulsate…"
(Williams 4th ed 1917, p342)

"As soon as the lungs begin to function, the circulation through the umbilical arteries normally ceases in from five to fifteen minutes after birth." (Williams 6th ed 1931, p418)

By the 1940s a change of opinion is evident:
"We have adopted an intermediate course, feeling that to always wait for complete cessation of pulsation frequently interferes with the proper conduct of the third stage of labor, and at the same time, that most of the available blood in the cord had been incorporated in the fetal circulation during the few minutes immediately following delivery." (Stander [Williams 8th ed] 1941, p429)

"Whenever possible, clamping or ligating the umbilical cord should be deferred until its pulsations wane or, at least, for one or two minutes…There has been a tendency of late, for a number of reasons, to ignore this precept. In the first place the widespread use of analgesic drugs in labor has resulted in a number of infants whose respiratory efforts are sluggish at birth and whom the obstetrician wishes to turn over immediately to an assistant for aspiration of mucus, and if necessary, resuscitation. This readily leads to the habit of clamping all cords promptly." (Eastman [Williams 10th ed] 1950, p397)

Would Williams recognize the 20th edition of his textbook?
"Although the theoretical risk of circulatory overloading from gross hypervolemia is formidable, especially in preterm and growth-retarded infants, addition of placental blood to the otherwise normal infant's circulation ordinarily does not cause difficulty… Our policy is to clamp the cord after first thoroughly clearing the infant's airway, all of which usually takes about 30 seconds."
(Cunningham et al. [Williams 20th ed] 1997, p336)
[Top]

7 - Developmental Delay
Monkeys subjected to asphyxia did not develop cerebral palsy as had been expected. However, monkeys asphyxiated at birth displayed transient difficulties with motor control.

The loss of neurons evident in figures 2 and 4 above would suggest that sound transmission to the temporal lobes might be totally blocked. However, the monkeys with damage in the inferior colliculi startled at the sound of buzzers, but they did not localize the sound as normal infant monkeys did [45, 46]. Sound localization has been shown to be a major function of the inferior colliculus [66-68]. The startle reaction may be a reflex response mediated by nuclei in the lower brainstem auditory pathway.
[Top]

8 – Poor Manual Dexterity
Figure 8 is a letter home from my son Ralf when he was sixteen and at a school for children with learning disabilities (or what Windle referred to as minimal cerebral dysfunction). The handwriting illustrates his limitation in manual dexterity – compared for example with that of a college-bound teenager. Large handwriting (macrographia) has been described in individuals with high-functioning autism [69].

Ralf suffered a traumatic birth and his development, including speech, was alarmingly delayed. We were told not to worry, "He's a boy. Boys are often slow," etc.

The rest of the world could care less about Ralf's "minimal" handicaps. I am painfully aware that he is not quite the person he would have been. Mostly he is not highly motivated or ambitious, but I know few people as sensitive to others, cheerful, and intent upon enjoying life. A few years ago I saw his "minimal cerebral dysfunction" recorded as PDD NOS. I dislike this label because of the pernicious implication that he and our family carry genes for "autism spectrum" disorder. Monkeys subjected to asphyxia at birth carried no such genes.

Ralf was crying at birth before anyone could answer my question whether the baby was a boy or a girl. A minute or so later the doctor held him up by his feet and announced, "It's a boy!" Ralf was howling and totally pink except for his feet, which were very blue. I suspect that oxygen deficiency occurred during my long difficult labor, and during which he incurred a prominent cephalhematoma. Because Ralf had a large head and forceps were required, I am sure a few minutes of total asphyxia occurred.

Conrad suffered a much more severe asphyxia; he was flaccid and ashen white at birth, and required resuscitation. I am sure Ralf suffered some impairment of function in the inferior colliculi, and that the same long-term progressive damage found in monkeys is the reason he is, as the neurologist told me many years ago, "not quite the person he would have been."

9 - Progressive Degeneration
Signs of on-going progressive neuropathologic changes were observed in monkeys kept alive for months or years following asphyxia at birth, even in monkeys without the characteristic lesions of the inferior colliculi [30]. Asphyxia had to be of seven to eight minutes duration before visible damage of the inferior colliculi was seen. Impairment of function within the inferior colliculus could reasonably be expected in cases where asphyxia was not of long enough duration to produce visible lesions.

Neurotransmitters are produced in the inferior and superior colliculi of immature laboratory animals, which are thought to guide formation of synapses and promote growth of later developing areas of the cerebral cortex [70, 71]. Submicroscopic as well as visible damage within the inferior colliculi might then disable the biochemical mechanisms required for normal maturation of the cerebral cortex. Failure to fully develop frontal lobe cognitive abilities becomes more and more painfully evident as a child with autism or "minimal cerebral dysfunction" matures. [Top]

Neuropathology found following long-term survival after asphyxia at birth involved:

Anomalies in many of these areas have been reported as part of the neuropathology of autism or its behavioral handicaps [72-82]. Functional impairment of the amygdala, cerebellum, hippocampus, and frontal lobe connections are at the forefront of theories of autistic disorder. Long-term brain changes following asphyxia at birth provide a mechanism by which impairment of these brain areas might come about, without having to hunt for gene loci on chromosomes.
[Top]

10 - Autism and complications at birth
Autism is not generally acknowledged as having anything to do with difficulties at birth. Perinatal problems have been discussed in several papers, but then usually dismissed as minimal or non-specific [57, 83-95]. As Juul-Dam et al. (2001) noted "No presently apparent unifying feature" can be identified as a unique predisposition for autism.

But oxygen insufficiency is the major "unifying feature" of all complications at birth. Respiration is the most immediate and essential need of all life forms dependent upon oxygen. Asphyxia at birth results in selective damage within the brainstem auditory pathway, and auditory dysfunction is a characteristic of autistic disorder. Failure to learn language by ear, as normal children all do, is the most serious obstacle to further development for children with autism. Asphyxia at birth warrants investigation as a cause of autism as much as if not before consideration of genetics, infections, or exposure to toxic substances.

Some researchers have argued that complications at birth are due to a pre-existing problem with the infant or mother. But damage to the inferior colliculi had nothing to do with any pre-existing problem of monkeys subjected to asphyxia at birth.

But the article by Willams et al. is a prime example of the difficulty finding visible signs of brain impairment: Reduced Purkinje cell density was noted in the cerebellum of a male patient who had a chronic seizure disorder and died at age twelve. This patient also had reduced pyramidal cell dendrites in the midfrontal gyrus. The same pyramidal cell abnormality was seen in the brain of a 27-year old autistic subject who belatedly was found to have phenylketonuria (PKU), and this was the only abnormality found!

Figure 9 is a drawing from an article by Towbin (1970) in a textbook of neuropathology [96]. The drawing bears such a striking resemblance to my son Ralf that I contacted Dr. Towbin, who told me the drawing was made in September 1962 in the newborn nursery at the Boston Lying-In Hospital. That is when and where Ralf was born, and if figure 9 is not a drawing of him, it looks just like him, even today in adulthood. His eyes were swollen shut right after birth (more so than shown in the picture), and he still bears a scar under his right eye from use of forceps.

I was told my difficult delivery with Ralf was because he was a "brow presentation." The article by Towbin and many chapters in textbooks of obstetrics describe the dangers of malpresentation at birth.
[Top]

11 - Mercury and Other Toxic Factors
Windle and coworkers found that bilirubin levels (from jaundice) can be very high and not cross the blood brain barrier (BBB), and therefore not damage the brain. In experiments with newborn monkeys, bilirubin was found to enter brain tissue only in those animals also subjected to asphyxia [54]. Compromise of the BBB by asphyxia may be part of a protective response to increase blood flow and help get oxygen into neurons.

Figure 10 shows staining by bilirubin of nuclei in the basal ganglia (subcortical motor nuclei) in a monkey subjected to asphyxia at birth. How can it be safe to assume that the immature brain is more resistant to oxygen insufficiency? Changes in the blood-brain barrier caused by asphyxia are not visible, and disruption of metabolic systems that precede the appearance of visible damage may still lead to serious impairment of function.

[Top]
II. THE AUDITORY SYSTEM

12 - Metabolic Rank Order
Myers (1972) described involvement of a "monotonous rank order" of brainstem nuclei in the pattern of damage caused by asphyxia at birth. This rank order is comparable to the metabolic rank order of brainstem nuclei revealed by the autoradiographic techniques for measuring cerebral blood flow and metabolism.

The autoradiogram picture in Figure 1 (top of webpage) is part of data gathered in experiments done nearly half a century ago (Landau et al. 1955) to investigate cerebral circulation [3]. A radioactive tracer was injected into a laboratory animal (cats were used in this initial investigation). Distribution of the tracer was measured one minute later.

Measurements of the aerobic enzymes alpha-ketoglutarate and cytochrome oxidase further confirm that high blood flow supports aerobic metabolism, which is highest in the same hierarchy of brain areas [12-14].

The auditory system is susceptible to injury because its components have greater metabolic needs than most other areas of the brain. On the other hand this most active system is clearly often spared. But after a few minutes of sudden total circulatory arrest, and if resuscitation is possible, the inferior colliculus incurs severe damage. This has been found true in adult human cases as well as children [111-117].
[Top]

13 - The Auditory System
Figure 11 is a diagram of the auditory system that shows the location of the darkly labeled structures in figure 1.

Roth and Barlow (1961) employed an autoradiographic technique modeled after that used for measuring blood flow to investigate distribution of drugs in the brain [122]. They found that the fast-acting anesthetic thiopental was quickly distributed to the inferior colliculus. Thiopental is used for rapid induction of general anesthesia (loss of consciousness). That thiopental goes directly to the inferior colliculi suggests that high metabolic rate in this pair of auditory nuclei may be important for maintaining the conscious state.

Deafness is not a handicap of consciousness or general awareness. But deafness is the result of impairment at the level of the cochlear nucleus, or mechanical components of the ears. Autism is a handicap of general awareness, or of multiple attention deficits at least. Lack of social awareness and diminished capacity for "shared attention" are manifestations of environmental obliviousness in children with autism. That these characteristics may stem from impairment of midbrain auditory alerting functions is worth exploring.
[Top]

14 - Auditory Dysfunction
That autistic children are hypersensitive, hyper-reactive, and confused by some sounds is common knowledge (though largely based on anecdotal accounts). Normal children carry on conversations in school cafeterias and gymnasiums as easily as anywhere else, even with music blaring from the loudspeakers. An autistic child in such a noisy setting may exhibit extreme distress. My son Conrad would cover his ears and often refuse to go into a room where he saw a telephone – another anecdotal account. But anecdotes should be collected and taken more seriously as sources of useful data.

A parent on an internet email exchange (autism@list.feat.org, 3 Oct 2002) asked for advice on what to do about classroom aides who were controlling his 13-year-old son's behavior by holding him down in front of a vacuum cleaner. An outpouring of sympathy, outrage, similar stories, and ways to help overcome the child's fear followed. Kanner (1943) described two of the eleven children in his original report as being afraid of the vacuum cleaner, one so much that she would not go near the closet where it was kept [123]. Analysis of sound patterns emitted by vacuum cleaners, ability to recognize words and other sounds with component sounds as background noise, and research on auditory evoked potentials to signals of interest presented in noisy surroundings might yield useful data.

Research to date on auditory evoked potentials suggests that acoustic signals from ear to temporal lobes may be slowed or distorted in some children with autism [124-134]. These investigations have been controversial but they do provide indication of auditory dysfunction. Further, monkeys asphyxiated at birth were not deaf, but measurement of auditory evoked potentials revealed a delay in auditory signal transmission similar to that found in some children with autism [135].

Confusion in noisy environments points to problems processing multiple incoming sounds, and suggests that alternatives to simple click and tone stimuli should be used in testing for disorders of hearing. For example, tests of word recognition in quiet (WRIQ) and word recognition in noise (WRIN) described by Church et al. (1997) could be used to assess verbal children and even adapted for use with low functioning children with autism [136].

Researchers at the molecular level have found that inhibitory as well as excitatory neurotransmitters work together to modulate responses of neurons that detect sound onset; ongoing signals of the same frequency and intensity are detected but not transmitted further [137, 138]. The hypersensitivity to sounds displayed by some autistic children may represent loss of inhibitory function – why the sound of a vacuum cleaner might be distressing beyond the imagination of most of us. Inability to distinguish sound onset then relegate it to background awareness could also be part of the difficulty in recognizing boundaries between words and syllables in spoken language.

Caspary et al (1995) provided data showing decline with advancing age of neurotransmitter function in the inferior colliculus that may lead to loss of the capacity to detect and extract meaningful signals from background noise [139]. They pointed out that this leads to difficulty following a conversation in a noisy environment and may be the reason some elderly people withdraw from participation in society. The same or similar disability may lead children with autism to avoid social contact.
[Top]

III. LANGUAGE

15 - Language by Ear
Language evolved in the human brain with the temporal-to-frontal lobe circuit in the cerebral cortex. Language is a higher cortical function. However, language learning begins before maturation of the cortical language areas is complete. Children learn language "by ear."

Detection of stressed syllables leads to a predictable first stage of language development known as "telegraphic speech" in which a child makes use of single syllable units of meaning and rearranges these newly acquired units of meaning to fit new contexts [143-145].

Inability to recognize syllable and word boundaries has been identified as a problem in some children with autism [146]. Use of phrase fragments rather than syllabic units is characteristic of the echolalic speech of autistic children. Phrase fragments are not as easily rearranged as syllabic units to fit new contexts.
[Top]

16 - Verbal Auditory Agnosia
Everyone agrees learning a language by ear is best, but after the first decade of life this becomes more and more difficult. Normal young children can learn a second language, without accent. The ability to hear syllable and word boundaries decreases with age. There is evidence that the auditory system ages faster than other areas of the brain because of its high metabolic rate [147]. Impairment of the inferior colliculus by asphyxia at birth may be like premature aging, and make it as difficult for a child so handicapped to detect syllable and word boundaries as it is for an adult trying to learn a foreign language.

Rapin (1997) suggested that inability to distinguish syllable and word boundaries in rapid streams of speech may be the basis of the language disorder in some children with autism; she referred to this as "verbal auditory agnosia" [146]. Agnosia is a failure of recognition without loss of sensory function. Most children with autism have hearing seemingly adequate for learning language, but agnosia ought to be more fully explored.

Agnosia for speech or "word deafness" (loss of ability to comprehend spoken language) has been reported in three previously normal adults following injury of the inferior colliculi in the midbrain auditory pathway [148-150].

17 - Echolalic Speech
Telegraphic speech includes use of elemental units of meaning rearranged to fit new contexts. Some examples are phrases like "Mommy go store" or "Apple, I want." Use of elemental units of meaning (morphemes) reveals a child's ability to parse what he hears. Command of morphemic components facilitates re-wording and comprehension of grammatical structures.

Speech productions of an autistic child on the other hand at best consist of phrase fragments used badly out of context. This is echolalic speech, use of parroted "sound bytes," which Kanner (1946) referred to as "irrelevant and metaphorical language" [151]. No rewording to fit new contexts takes place. "Pronoun reversal" is part of the failure to re-word, as is frequent use of the prosodic intonation for a question.

"You don't want to go swimming?" is not intended as a question, but as a statement by the child that he doesn't want to go in the water. "You want an apple?" is equivalent to the telegraphic request "Apple, I want." "Is that yours?" is equivalent to a normal young child's statement, "That mine."
[Top]

IV. CHILDHOOD HANDICAPS

19 - Auditory and Motor Handicaps
The syndrome of autism described by Kanner (1943) did not include indications of neurological problems [123]. Children with the "core syndrome" of Kanner autism develop motor milestones on time, are toilet trained on time, and even appear to begin speaking on time. By the age of three or four however, speech development is clearly aberrant.

Delayed motor development and residual physical awkwardness are however indications of neurological impairment in many children with autism, and especially in those with a diagnosis of Asperger's syndrome. Few children fit the narrow description of Kanner autism; Kanner even commented on the individual differences he observed. To explain the autism spectrum: Impairment of function in the inferior colliculi (by asphyxia at birth) might be at one end of the spectrum and involvement of motor systems (by hypoxic birth) at the other. Most children with autism show signs of both auditory dysfunction and some delay of motor development.

Asperger's syndrome is a far more hopeful diagnosis than autism. Children with Asperger's syndrome may be late learning to speak and their speech is full of pedantic sound-bytes with often peculiar and pun-like connections to conversational context. But command of language is the tool by which children with Asperger's syndrome can learn and grow. Language disorder is the most serious impediment to human development.

Lack of manual dexterity was noted as the most common residual deficit in monkeys asphyxiated at birth. The normal climbing ability of monkeys was also never achieved. Control of wrist, ankles, and digits remained inadequate. A reduced level of spontaneous activity was observed. The asphyxiated monkeys were described as hypoactive, docile, unemotional, and not easily disturbed.

Short-term memory also appeared deficient in asphyxiated monkeys. But tests of learning involved repeated trials, and the asphyxiated monkeys were described as difficult to coax as if they gave up shortly after beginning training trials. The behaviors were compared to the condition known then as "minimal cerebral dysfunction" (MCD) in children. Clinical signs listed for MCD included problems with attention, impulse control, interpersonal relations, and hyper- or hypo-reactivity, along with lack of coordination and learning disabilities. Some children with MCD might now be viewed as having Asperger's syndrome or attention-deficit hyperactivity disorder (ADHD).
[Top]

20 - Increased Incidence of Autism
As long ago as 1975 I proposed that echolalic speech of children with autism might be the result of damage to the inferior colliculi, and that asphyxia at birth is an example of how such damage could take place [152]. But now immediate clamping of the umbilical cord has become a standard obstetric practice, whether or not the infant has begun to breathe. Unless breathing is established within seconds after cutting off umbilical circulation, asphyxia will occur and brainstem nuclei like the inferior colliculus will be damaged or impaired.

Four minutes is generally recognized as the time beyond which resuscitation becomes less likely following accidents such as drowning or cardiac arrest. It is a common belief that infant humans and animals can withstand oxygen deprivation longer than mature beings. But Myers (1972) described changes in neurons (seen only under the microscope) in the inferior colliculus of monkeys subjected to asphyxia of duration too short to cause visible damage [2]. Asphyxia had to be of seven to eight minutes duration before visible damage was seen, and Faro and Windle (1969) found progressive neuropathologic changes in monkeys kept alive for months or years following asphyxia at birth, even in monkeys without the characteristic lesions of the inferior colliculi [30].

It would seem dangerous to assume that a newborn child is more resistant to anoxia and can withstand a few minutes of asphyxia until breathing is initiated by artificial ventilation. Myers found that it is the infant heart that withstands asphyxia better than that of the adult (not the brain). Failure of attempts for prompt resuscitation after cutting the cord will result in at least some impairment of the brain.

Who has the evidence that minimal brainstem damage is not important? Research data clearly implies that damage confined to the brainstem results in developmental deficits. That early brain damage heals because of "plasticity" was disproved when Faro and Windle discovered that with time early brainstem damage leads to progressive widespread changes throughout the brain.

Delayed onset of breathing after the cord has been cut deserves consideration as a contributing factor to increased numbers of children with developmental disorders, and the increased incidence (or prevalence) of autism. A low Apgar score five minutes after birth is considered ominous, and has been correlated with later developing autism [56, 57]. Increases in autism have been noted during the same period that immediate cord clamping has become routine.

In addition to complications at birth, autism is associated with many medical conditions, which include prenatal exposure to alcohol and drugs, pre- and postnatal infections, lead poisoning, gastrointestinal disorders, neurologic (seizure) disorders, and diverse genetic predispositions [57, 83-95,153-179, 180-210]. All of these conditions are likely to have a catastrophic effect on metabolism similar to that caused by asphyxia at birth. Prenatal exposure to alcohol, drugs, and medications might be linked to the increased prevalence of autism. But genetic mutations are unlikely to be increasing at such a high rate.
[Top]

21 - Fetal to Postnatal Adaptation
Mercer and Skovgaard (2002) noted that early clamping of the umbilical cord may have been put into practice without adequate evidence of its safety [211]. They further described how clamping of the cord obstructs normal completion of transition from pre- to postnatal life. Most dangerous is the potential reduction of neonatal blood volume from 25 to 40 percent.

Ventilation alone is not sufficient to expand the lungs. Mercer and Skovgaard call attention to the need for adequate blood volume to stimulate erection of capillaries in the alveoli of the lungs, and thus to initiate oxygen transfer to the red blood cells. They point out also that activity of other organs, like the gut, is low during gestation, and capillary erection may also be involved in stimulating function of all body organs.

In the fetal state 40 percent of the cardiac output is to the placenta to obtain oxygen from the mother. Transition to postnatal life depends upon utilization of blood from the placental circuit to activate all organs for extra-uterine survival. In the fetal state the lungs secrete amniotic fluid, and according to Mercer and Skovgaard, "capillary erection may be the natural stimulus for the lung to change both structure and function immediately at birth from an organ of fluid secretion to an organ of gas exchange."

If an infant cries immediately at birth this transition has at least for the most part occurred. But it appears that the child who does not cry right away is the first to have the umbilical cord cut and be taken away for ventilation and other desperate efforts (including artificial blood volume expanders) to initiate breathing! These are the infants who could most benefit from allowing placental circulation to continue.

Mercer and Skovgaard note that, "Since the beginning of mammalian life, young have been born attached to a life-line that supports their transition to extrauterine life." They call attention to two exceptions: (1) human birth in recent times, and (2) attended births of some thoroughbred foals, which included rapid clamping of the umbilical cord, and in which a "convulsive syndrome" often occurred. Normally a mare and foal rest for about half an hour after birth, and the cord is broken only when either the mare or the foal rises.

Pathology found in foals that died of the convulsive syndrome included an absence of aeration of the alveoli. [Top]

22 - Forgotten History
Adequate blood volume is required at birth to stimulate opening of the capillaries within the lungs, gut, and other body organs that are dormant during gestation. But depriving an infant of placental blood at birth also leaves the child anemic. Wilson, Windle, and Alt (1941) investigated clamping of the umbilical cord as the cause of iron deficiency anemia in infancy [212]; they noted that the diet of an infant up to the end of the first year cannot make up for this deficiency. Infant anemia has since been found correlated with early childhood learning disorders [213-214].

It appears that during the 1930s use of anesthesia in childbirth and sluggish respiratory efforts of the newborn led to development of protocols for resuscitation, which included early clamping of the cord. The children described by Kanner in 1943 were all born during the 1930s, and at least two by cesarean. Anesthesia was first used in 1846, but obstetric textbooks through the 1930s still encouraged allowing the umbilical cord to cease pulsation before cutting it [58-65]. The teaching of older texts should be heeded, at least that the umbilical cord be left intact until the newborn infant is breathing on its own. How shocking that what was so clearly understood in the past has been disregarded and totally forgotten.

But the unifying feature of mild and nonspecific complications is the likelihood of lapses in oxygen delivery during the transition from placental to pulmonary respiration. In the hierarchy of human needs nothing is more essential and of immediate urgency [215].

Interference with respiration at birth should be thoroughly re-investigated and "ruled out" before any more esoteric causes of autism are entertained. The research of Windle and Myers still provides evidence that complications at birth can have serious consequences, and their findings merit continuing consideration in the search for understanding and preventing developmental disabilities.

23 - Worth Remembering
The following quotes from Windle (1969) are points worth remembering:

• "In any delivery it is important to keep the umbilical cord intact until the placenta has been delivered.
  
  To clamp the cord immediately is equivalent to subjecting the infant to a massive hemorrhage, because almost a fourth of the fetal blood is in the placental circuit at birth."
  
  
• "Spontaneous neurological deficits are practically unknown among rhesus monkeys born in their natural habitat…
  
  The female squats and drops the infant on the ground. During delivery most of the blood in the placenta passes to the infant."
  
  
• "It is no longer acceptable to assume that the human fetus or newborn infant is so resistant to oxygen deficiency that it will escape harm from a short exposure to asphyxia neonatorum.
  
  If the infant's brain can be compared to the monkey's, asphyxia of such duration that resuscitation was required will certainly have damaged it."
  
  
• "The briefly asphyxiated infant monkeys with minimal brain damage lost their signs of neurological deficit… The extent of this 'recovery' was surprising.
  
  The residual deficits of the surviving animals are now inadequate manual dexterity…"
  
  
• "It is commonly recognized that improvement can be expected after a distressful birth…
  
  We know that the brain of a 'recovered' monkey is structurally damaged, whereas we only assume on clinical grounds that the brain of a 'recovered' human infant is normal."

24 – Hemoglobin
Respiratory gas exchange is a biochemical process mediated by the hemoglobin molecule within red blood cells. The binding and release of oxygen from hemoglobin provides an explanation for why hypoxia is so different from the effect of asphyxia. The quickest adjustment to an environment of oxygen insufficiency is that provided by the action of hemoglobin in delivering oxygen first to tissues producing the most carbon dioxide.

Myers (1972) demonstrated that while the inferior colliculi are predictably and selectively damaged by a few minutes of total oxygen deprivation at birth, they are spared during a period of prolonged partial oxygen insufficiency. It takes a catastrophic and complete obstruction of aerobic metabolism for involvement of the inferior colliculus to take place. But this can happen to an infant born not breathing if the umbilical cord is cut before respiration can be established.

Survival of multi-cellular organisms depended upon evolution of a more efficient means for exchange of respiratory gases. As White et al (1969) commented, the active metabolism of mammalian tissues remote from the atmosphere is possible only because, "Through the action of hemoglobin, oxygen is abstracted from the air, carried within a few seconds to the most distant parts of the body, and delivered to the tissues at a pressure only slightly less than that at which it existed in the atmosphere" [216].

25 – Infant Anemia
Windle warned that clamping the umbilical cord immediately after birth is equivalent to subjecting an infant to a massive hemorrhage. In 1941 a group of researchers including Windle reported iron deficiency anemia (low hemoglobin levels) in infants at 8 to 10 months of age whose umbilical cords had been clamped immediately at birth [212]. Hemoglobin is the iron-containing molecule of red blood cells; a deficiency of iron implies a deficiency of hemoglobin upon which oxygen delivery depends.

Deficient hemoglobin is equivalent to a hypoxic environment; less oxygen will be available for growth and development of the brain and other organs. Anemia in infancy is a state of chronic partial hypoxia, the effects of which can be compared to those produced by Myers (1972) on partial obstruction of umbilical blood flow.

Perhaps some of the on-going degeneration within the brains of monkeys asphyxiated at birth was due to anemia resulting from the way asphyxia was imposed – by clamping the umbilical cord. The residual inadequate manual dexterity of monkeys asphyxiated by umbilical cord clamping at birth could as well be the result of involvement of motor areas of the brain from postnatal anemia as from brainstem damage.

The paper by Saigal and Usher (1977) appears to have initiated the fear that delayed clamping of the umbilical cord could result in polycythemia (too many red blood cells) and jaundice [47]. But polycythemia is a physiological response to abnormalities like methemoglobinemia, which results from a genetic or drug-induced abnormality of the hemoglobin molecule [217, 218-220]. It may be time to question the opinions of modern authorities and look back again at some forgotten history.

Jellett (1910) in his Manual of Midwifery discussed the issue of polycythemia after stating, "The old dispute as to when the cord should be tied possesses now little more than an academic interest, as it is conclusively settled that this should not be done until all pulsations in the cord have ceased" [221].

Jellett cited research known at that time (but long since forgotten). White (1785) had written about the absurdity of supposing that it was possible for the change from placental to pulmonary circulation, with all that this implies, to take place in a moment, "that this wonderful alteration in the human machine should be brought about in one instant of time, and at the will of a bystander?" [222].

Jellett further cited research by Schmidt (1894) in which he found that 72 percent of children in whom immediate ligation of the cord was done were jaundiced, while only 42 percent were jaundiced when the cord was not tied until ten minutes after birth. It may be time to consider whether postnatal anemia isn't a greater risk for more infants than polycythemia and jaundice [223].

26 – Autism in Twins
Autism is not 100 percent concordant in identical twins. The finding of even one pair of identical twins who are discordant for any disorder provides the counterexample that disproves a simple genetic etiology.

Folstein and Rutter (1977) investigated cases of autism occurring in twins [224]. Eleven pairs were identical (monozygotic), and ten fraternal (dizygotic). Concordance for autism was found in 4 of the 11pairs of monozygotic twins (36 percent) and no concordance was found in the dizygotic pairs. Thus of 21 twin pairs, 17 were discordant for autism; and in 12 of these autism was associated with an event likely to cause brain damage. Case reports are provided and worth reviewing:

• Of the identical twin pairs concordant for autism, all four are male and complications of pregnancy were noted in each case. For example, the mother of one pair went into labor of 24 hours duration six weeks early, and each twin was a breech birth. Another mother of a concordant pair had labor induced at 39 weeks gestation because of pre-eclamptic toxemia; the second twin was born 30 minutes after the first due to uterine inertia. He suffered fetal distress, and did not breathe until 7 minutes after birth; autism and cognitive disability were more severe in this second-born twin than in his brother. The differences are more striking than the similarities in each of the four twin pairs deemed concordant for autism.
• Three of the five identical twins discordant for autism were female. One of the female twins who became autistic was a breech birth with delayed breathing; her umbilical cord was described as very narrow and white. One of the male twins with autism also had a cleft palate, which suggests prenatal exposure to alcohol or other drugs.
• Of the non-identical twin pairs, three of the mothers were Rh-negative. The three twin pairs of these mothers were all male. The only twins concordant for cognitive disorder were born to one of these mothers; she did not have Rh-factor antibodies but bilirubin rose to 10 mg in the neonatal period of the twin who became autistic. Exchange transfusions were performed in both twins of one Rh-negative mother. Of the third Rh-negative mother, only the twin who later became autistic had an exchange transfusion at birth.

It is difficult not to question the role of perinatal compromise in all of the cases described by Folstein and Rutter. The most obvious genetic predisposition is the occurrence of Rh-negative blood type in three of the mothers.
[Top]

Norman (1982) noted that perinatal hazards are increased for twins and suggested that therefore twins are an imperfect model for genetic versus environmental studies of things like intelligence [225]. That concordance is higher in identical than fraternal twins reflects factors such as the limited capacity of a shared placenta to withstand environmental hazards like anoxia or prenatal infections. Davis et al. (1995) found that twins who both develop schizophrenia were more likely to have shared a single placenta and chorionic sack in utero [226]. Autism becomes evident years earlier than schizophrenic disorders and is therefore even more likely related to prenatal environment, and autism may yet prove to be part of the schizophrenia spectrum

Ritvo et al. (1985) reported concordance for autism in 22 of 23 identical twin pairs (95.7 percent) and in 4 of 17 fraternal twin pairs (23.5 percent) [227]. The discordant identical twins were male, age 8, with a normal sister, one year younger. Of the 22 pairs of identical twins concordant for autism, 5 were female, comparable to the 4:1 ratio of males to females reported for autism occurring in the general population.

In the study of Ritvo et al., concordance for autism among the 17 pairs of fraternal twins was high, occurring in 4 pairs. Of the 13 pairs of fraternal twins discordant for autism, 5 were male-female twins and the autistic twin was male in 2 pairs. Both of the male-female twin pairs, in which the male was autistic, had male siblings who were autistic. Of the 8 same-sex twin pairs discordant for autism, only one was female.

Steffenburg et al. (1989) investigated occurrence of autism in same-sexed twins under the age of 25 [85]. They found 21 twin pairs, 11 monozygotic (including one set of identical triplets) and 10 dizygotic. Concordance for autism was 91 percent in the monozygotic pairs. Concordance for autism was not found in the dizygotic pairs, but concordance for cognitive disorder was 30 percent. In the twin pairs discordant for autism, autism was associated with greater perinatal stress. Steffenburg et al. concluded that autism sometimes has a hereditary component, and that perinatal stress is involved in some cases. Case reports were not provided.

Greenberg et al (2001) noted a higher concordance of autism in fraternal twins than would be expected in the general population, and this indicates environmental influences are more significant than genetic factors [93]. If environmental factors were not involved, the concordance rate for fraternal twins should not be greater than between single-born siblings in families in which autism has occurred more than once.

Autism has been found as a complication of phenylketonuria, and other genetic disorders. But some of the medical conditions associated with autism may have been mistakenly thought of as genetic. For example Migeon et al. (1995) and Subramaniam et al. (1997) described a pair of identical twin girls in which one had Rett syndrome but the other was developing normally still at the age of six [228, 229]; and Feekery et al. (1993) reported Landau-Kleffner syndrome in one but not the other of identical twins [230].

27 – Male-Female Differences
In my own dissertation research, I with my advisor, Ladislav Volicer, investigated the long-term effects of neonatal asphyxia in laboratory rats. The major findings of this research have been published (Simon and Volicer 1976) [231].

Newborn rat pups were subjected to asphyxia by suffocation in small air-tight vials until gasping efforts ceased (45 minutes to two hours). Pups were pale and flaccid when removed from the vials; they were resuscitated using a slow stream of air to the nose and mouth and massaging the chest. Only about half of the experimental animals survived and they were lethargic during the first 24 hours following resuscitation. Most did not gain the normal amount of weight and many lost weight during the first 24 hours; animals with weight loss exceeding 0.5 to 1.0 grams or that appeared jaundiced (with yellow discoloration of the skin) often died during the first two days. Control animals gained about one gram during the first 24 hours after birth.

But growth retardation persisted for the first two weeks of life, most noticeably in male rats. Figure 17 is a graph showing differences in weight between pairs of male and pairs of female animals. Brain growth was also retarded in the asphyxiated rats, and to the same degree in both males and females.
[Top]

We thought that vulnerability to asphyxia might be related to the metabolic requirements of the animal and that this might be related to birth weight and sex. Male animals were significantly heavier at birth (just over 7 grams) than females (just under 7 grams), with standard deviation 0.07 grams. Metabolism would appear to be higher in males than females. Research evidence is not abundant on this subject, but it is common knowledge that males (on average) have greater muscle mass than females and have greater muscular strength. Almost all competitive athletic events have male and female categories, or female champions and record-setters would be few and far between if at all.

Growth retardation of male laboratory rats indicates they were more severely affected by asphyxia. The inferior colliculi and other brainstem nuclei are likely affected sooner in males than females; thus impairment should occur with a shorter period of asphyxia.

Again, before complex hypotheses are investigated in search for the brain disorder in autism, the most basic requirement for aerobic organisms deserves thorough study. Damage of the auditory system by asphyxia at birth is worth further research as cause of developmental language disorder.
[Top]

V. BRAINSTEM DAMAGE

28 - Variable Vulnerability
Preservation of function in the auditory system is apparently important for survival and homeostasis. Mechanisms that increase blood flow under adverse conditions protect the auditory system in most circumstances, as shown by research using the autoradiographic methods for blood flow and deoxyglucose uptake For example the nerve gas Soman administered to laboratory animals in sub-toxic doses stimulated increased blood flow, with the greatest increase in the inferior colliculus [233-235].

Adjustment of aerobic metabolism as well as increased blood flow takes place with use of alcohol and other drugs. Research on the effects of amphetamine drugs revealed increased blood flow and metabolism in the inferior colliculus and visual cortex with metabolic depression elsewhere in the brain [236].

Grünwald et al. (1993) found that administration of one dose of alcohol to laboratory rats depressed glucose utilization throughout the brain, and this was most pronounced in the auditory system [24]. Increasing doses of alcohol given over a three week period increased deoxyglucose uptake in the inferior colliculus in dramatic contrast to lowered uptake in other areas of the brain. But an additional dose given two hours before measurement of deoxyglucose led to depressed uptake in the inferior colliculus.

If increased metabolism leads to auditory sensitivity that is diminished by an additional dose of alcohol, this might provide one explanation for addiction. In any event it provides an example of how metabolism as well as blood flow adapts to substances like alcohol.

Neuropathology caused by alcohol intoxication has long been known (as Wernicke's encephalopathy, in Wernicke-Korskoff syndrome), and the mammillary bodies are most predictably involved [237-241]. The mammillary bodies are among the brain areas of high glucose utilization as can be seen in table 2. Protective mechanisms that spare the inferior colliculi may then make slightly less active nuclei like the mammillary bodies most vulnerable to compromise.

Damage to the cerebral cortex is the most common consequence of oxygen deficiency. Clamping the carotid arteries of neonatal gerbils led to decreased blood flow to the forebrain but increased circulation in the inferior colliculi from collateral blood vessels [21]. Partial obstruction of placental blood flow to fetal monkeys produced damage in the cerebral cortex and cerebral palsy [2].
[Top]

29 - Patterns of Damage

(1) Cortical and brainstem patterns
Damage caused by partial oxygen insufficiency (hypoxia) and total oxygen cutoff (asphyxia) is not a matter of degree. That brainstem damage is sometimes seen in human cases of circulatory arrest or suffocation has been a puzzle and topic of discussion in the medical literature [111-117].

Neubuerger (1954) investigated neuropathology in cases of cardiac arrest during surgery; most resulted in damage to the cerebral cortex, but Neubuerger expressed surprise that the location of lesions showed more variety than expected [111]. He made special note of one case in which there was, "...involvement of centers usually affected in Wernicke's disease, especially the mammillary bodies and posterior colliculi..." This was a person who died, at age 72, of respiratory failure during surgery. Artificial respiration and heart massage were performed with spontaneous heart beats obtained after eight minutes and spontaeous respiration after eighteen minutes. The patient, however, remained in a coma until death one week later. This can be viewed as a case of sudden catastrophic asphyxia, comparable to asphyxia at birth as produced in the experiments begun five years later by Ranck and Windle 1959 [26].

Gilles (1963) described symmetrical damage of brainstem nuclei in three individuals who survived episodes of cardiac arrest but later succumbed [113]. In two of the cases, normally expected damage was found in the cerebral cortex, thalamus, and cerebellar cortex. In addition, brainstem lesions of sensory and motor nuclei of several cranial nerves including the third (oculomotor), the inferior colliculi, cell groups in the reticular formation, and the lateral cuneate nuclei were found. The third case was an eighteen-month-old child who drowned but was resuscitated. In this case, the cerebral and cerebellar cortex remained intact.

Gilles noted the similarity of the brainstem pattern of damage to that found after experimental asphyxia of newborn monkeys and in human cases of kernicterus (bilirubin staining, or jaundice, of subcortical nuclei), and he suggested that brainstem "aphasias," such as in Moebius syndrome [242] might be related to temporary cardiac arrest in the prenatal or perinatal period. Moebius syndrome is another condition associated with Autism [243, 244], and it appears to be related to brainstem pathology [245-248], caused in some cases by prenatal exposure to the drug misoprostol [249].
[Top]

(2) Wernicke-like or "circulatory arrest" encephalopathy
Brierley (1961) described, "Wernicke-like" lesions in the mammillary bodies as well as severe cell loss and gliosis in the inferior colliculi" in the brain of a two-year old child who suffered cardiac arrest of ten to fifteen minutes duration from anesthesia used for repair of a hernia [112]. Brierley also noted the striking similarity of this pattern of damage to that reported by Ranck and Windle.

Gilles (1969) reported three more cases of brainstem damage: In a 17-month-old boy who suffered cardiac arrest during a high fever, a nine-year-old child resuscitated after suffocation under a collapsed earthen bank, and a 22-year-old man who suffered cardiac arrest following a bullet wound to the abdomen [116]. Gilles referred to these cases as examples of "hypotensive brainstem necrosis." Janzer and Friede (1980) suggested "cardiac arrest encephalopathy" as a better term, and reported the brainstem pattern of damage in eight infants and seven adults for all of whom a well documented episode of cardiac arrest had occurred [117].

When circulation and oxygen delivery are completely cut off, metabolism within the brainstem nuclei of high circulatory rate comes to a sudden halt. This clearly happens far less often with survival than circulatory insufficiency or partial disruption of aerobic metabolism. Damage from both partial and total asphyxia can happen, with both cortical and brainstem patterns evident in the brain of any victim of an accident that interferes with respiration.

Asphyxia is not a more severe degree of hypoxia; asphyxia is a different kind of insult. Protective mechanisms go into effect in situations of reduced blood flow or oxygen insufficiency (hypoxia). Expansion of blood vessels (vasodilation) happens with exposure to toxic substances [233-236], and in alcohol intoxication [24]. Increased blood pressure can lead to bursting of capillaries; the "whiskey nose" of some alcoholics is a visible sign of this kind of process. Figure 19 (below) shows "flea-bite" size hemorrhages in the brainstem caused by alcohol intoxication. The "corpus quadrigeminum post" (posterior quadrigeminal body, or inferior colliculus) is indicated as a primary site in which this kind of damage occurs.
[Top]

30 – Wernicke's Encephalopathy

(1) Hemorrhagic (or vascular) cause?
Brainstem damage caused by alcohol intoxication has long been known as Wernicke's encephalopathy [237-241]. Wernicke (1881) observed this pattern of damage in the brains of two alcoholic men and a young woman who had swallowed sulfuric acid [251]. An English translation of part of Wernicke's paper can be found in the article by Brody and Wilkins (1968) [253].

Note the involvement of the cerebellum in Wernicke's encephalopathy [254, 255]. Similar cerebellar pathology is prominent in the few investigations of brains from individuals with autism.

Damage caused by asphyxia at birth should probably be recognized as a variant of Wernicke's encephalopathy. Windle reported that the brainstem lesions caused by asphyxia at birth were not hemorrhagic in origin, which may be why he did not consider the damage caused by asphyxia at birth to be a variant of Wernicke's encephalopathy. But the brainstem structures affected are comparable.
[Top]

31 - Suffocation at the Molecular Level
Circulatory arrest and suffocation have an immediate and catastrophic effect on respiratory gas exchange and thence aerobic metabolism. In these circumstances brain structures of high metabolic rate are most vulnerable, evident from experiments on asphyxia in monkeys (newborn and adult) and case reports of damage following resuscitation after cardiac arrest and accidents like drowning [1-2, 29, 111-117]. The same metabolically active brain nuclei are also susceptible to damage or impaired function by chemical substances many of which may selectively disable aerobic metabolic pathways.

(1) Selective disruption of aerobic metabolism:
Troncoso et al. (1981) introduced the substance pyrithiamine as a tool for simulating thiamine deficiency, producing Wernicke's encephalopathy, and as a shortcut for mimicking the effects of chronic alcohol consumption in laboratory animals [256]. Pyrithiamine displaces thiamine (vitamin B1) from the enzyme that activates it as a cofactor for the major (Krebs cycle) pathway of glucose metabolism. Neurological impairment and neuropathology are produced more quickly with pyrithiamine than by administering alcohol to experimental animals [257-261].

Pyrithiamine causes suffocation at the molecular level. Protective mechanisms can have no effect. Hakim (1986) found dramatic increases in cerebral circulation following pyrithiamine administration – a protective response doomed to failure [259]. Blood flow returned to normal levels following administration of thiamine except in the inferior colliculi, mammillary bodies, and thalamic nuclei – an early sign of permanent impairment in these metabolically most active brain nuclei.

Glucose metabolism is blocked by pyrithiamine, thus carbon dioxide is not produced as a stimulus for hemoglobin to release oxygen. Chen et al. (1997) found the inferior colliculi sustained damage before the mammillary bodies or any other of the brainstem nuclei of high metabolic rate in laboratory rats injected with pyrithiamine [261]. The blocking of glucose metabolism by pyrithiamine is one step beyond that of carbon monoxide or cyanide, which displace oxygen on the hemoglobin molecule [33, 34].
[Top]

(2) Fast-acting anesthesia:
Roth and Barlow (1961) used an autoradiographic technique modeled after that used for measuring blood flow to investigate distribution of anesthetic agents in the brain [122]. They found distribution in the brain of fast-acting thiopental was different from that of barbiturates, which are slow to take effect but of longer duration. Thiopental was quickly distributed to the inferior colliculi and other nuclei of high blood flow. This finding lends support to the idea of Denny-Brown (1962) that the midbrain tectum (auditory and visual colliculi) is involved in maintaining the conscious state [119].

If the action of thiopental is on a specific neurotransmitter system, it would appear to be one of special importance for consciousness, and one closely coupled to aerobic energy production – the reason for high aerobic metabolism to have evolved in brain nuclei that maintain vigilance for environmental change.

(3) Brain-damaging intoxicants:
Damage of the inferior colliculus has been reported following exposure to several toxic chemicals [262-267]. Fumes from dry-cleaning fluids may reduce oxygen available to the lungs, or like thiopental be quickly distributed to areas of the brain with high blood flow. A confusional delirium afflicting two pressers in a dry-cleaning establishment was described by Bini and Bollea (1947) as comparable to alcohol intoxication [262]. Chronic intoxication lead to coma and death with focal lesions typical of Wernicke's encephalopathy found in the mammillary bodies, floor of the fourth ventricle, and inferior and superior colliculi.

Franken (1959) reported the case of a 28-year-old man exposed over a two-year period in his work filling fire extinguishers with methyl bromide [263]. He developed transient alterations in consciousness and involuntary jerking of his legs; he died two days after acute intoxication by a large amount of methyl bromide. Franken described brain lesions of the Wernicke-encephalitic type involving both posterior quadrigeminal bodies (the inferior colliculi). Goulon et sl. (1975) and Squier et al. (1992) reported additional cases of disability and death following acute exposure to methyl bromide, with Wernicke-like patterns with prominent involvement of the inferior colliculi [264, 265].

Cavanagh (1992) commented on the paper by Squier et al. and pointed out other toxic substances (6-amino-nicotinamide, misonidazole, metronidazole, and mono- and dinitro-benzene) that act by disabling the biochemical pathways of energy generation and produce symmetrical brainstem damage similar to that caused by methyl bromide [266]. Cavanagh and Nolan (1993) described the same pathology in laboratory rats caused by alpha-chlorohydrin, which was under investigation for use as a male anti-fertility agent [267]!
[Top]

(4) Selective what?
Terms such as "Selective serotonin re-uptake inhibitors" have become part of the common vernacular, and are used to promote sales of drugs touted to improve mental well-being. Caution is warranted. There may be drugs, and even herbal remedies, that impair aerobic metabolism without causing visible damage; unlike alpha-chlorohydrin they may be out there on the market.

Alzheimer dementia as well as autism is on the rise. It might seem prudent to require that all pharmaceutical products be tested for their effects on the brain. The autoradiographic methods for blood flow and deoxyglucose uptake are ideally suited to investigation of metabolic effects of substances that do not produce visible damage. Auditory system and aerobic metabolic impairments have been detected in Alzheimer patients [14, 268].

(5) Mercury and lead poisoning:
Bertoni and Sprenkle (1989) used the deoxyglucose method to investigate how lead compounds affect the brain. They found decreased glucose uptake throughout the brain seven hours after administration of lead acetate, but this was most pronounced in structures of the auditory system: Inferior colliculus, superior olive, cochlear nucleus, lateral lemniscus, and auditory cortex [18]. Bertoni and Sprenkle proposed that lead may poison the energy-activating enzyme Na,K-ATPase.

Oyanagi et al. (1989) reported auditory system involvement as part of the neuropathology of mercury poisoning in 14 victims of Minamata disease (or Hunter-Russell syndrome), a condition that includes impairment of hearing and speech [102]. The evidence that mercury and lead affect the auditory system suggests that they may be disruptive to aerobic metabolism or a neurotransmitter system dependent upon a steady energy supply. How heavy metals interfere with aerobic metabolism is not as clear-cut as in the case of pyrithiamine.

The mercury preservative (Thimerosol) in vaccines is widely claimed to be a cause of autism [97-101]. The mercury theory is controversial because it has been formulated largely by parents who report having seen regression into autism following vaccination of their child. Lead poisoning has been reported as a co-morbid condition in some children with autism[174-176].
[Top]

(6) Mitochondrial damage:
Aerobic metabolism takes place within the mitochondria of cells. Any substance disruptive to any component of mitochondria (cell walls, cristae, enzyme complexes, ribosomes, or DNA) will impede aerobic metabolism. Mitochondria have their own double stranded circular DNA similar to that of primitive bacteria [274, 276]. Mitochondria are thought possibly to have been a primeval infecting organism that turned out to be symbiotic for multicellular life forms because of the efficiency of their aerobic pathways for energy production [270, 277].

Mitochondrial genes encode RNA sequences for ribosomes, where peptides are assembled to become components of the enzymes that catalyze oxidative metabolism [274, 276]. Genes within chromosomes of cell nuclei encode additional RNA and peptide sub-units for the mitochondrial enzyme system. Mitochondria are duplicated from the maternal egg because sperm cells do not contain mitochondria [277]. Chromosomal genes are derived from both parents, and are far less vulnerable to mutations because protective enzymes that repair chromosomal DNA evolved later in cells with nuclei.

Mitrochondrial DNA is especially susceptible to mutagenic agents, and antibiotic drugs can damage mitochondria because of their similarity to bacterial organisms [271, 275]. Antibiotics are less and less effective against bacteria, which mutate and produce resistant strains. But damage to mitochondria may be a more urgent reason to avoid overuse of antibiotic medications. Mutations of mitochondrial DNA can lead to serious mitochondrial disorders. The kidneys and auditory system are most susceptible; nephrotoxicity and ototoxicity are common side effects of many drugs.

Matrilineal transmission of deafness induced by widespread use of streptomycin has been reported [272, 273]. Toxic chemicals used as herbicides and pesticides have long been known to cause neurodegenerative disorders; methyl bromide was used for this purpose as well as in fire extinguishers. DeJong (1944) reported the neurotoxicity of methyl bromide [269], which later was found to cause a brainstem pattern of damage similar to that caused by asphyxia [262-265]. Other chemical substances have been found to be mutagenic to mitochondrial DNA and are known to cause symptoms similar to those of Parkinsonism and Huntington’s chorea [271, 275]. [Top]

32 - Thiamine Deficiency
Suffocation at the molecular level also occurs in cases of severe thiamine (vitamin B1) deficiency. Beriberi is an illness that developed in countries where rice was a dietary staple after whole-grain rice was replaced by refined white rice [282, 284].

Causes ranging from infection, toxic contamination of water, dietary deficiency of protein, to racial predispositions were sought, but adding back to rice "polishings" containing the germ removed during refining was found to immediately ameliorate the symptoms of this illness. Diligent research finally led to isolation of thiamine as the essential molecular component that would prevent illness [281].

Thiamine is an essential cofactor for enzymes that catalyze glucose metabolism, and is now therefore classified as a vitamin. Thiamine is needed only in small amounts but it has a high turnover rate and needs to be replaced at least daily. Neurological signs gave evidence of brain involvement in beriberi. Peripheral "polyneuritis," disorders of eye movements, staggering gait, disorientation, and drowsiness were noted to be similar to the problems of chronic alcoholics, and thiamine deficiency was found to produce the same bilaterally symmetric lesions of brainstem nuclei characteristic of Wernicke's encephalopathy [283].

Figure 21 shows severe damage to the inferior colliculi in a terminally ill patient maintained on total parenteral nutrition (TPN) in which thiamine was lacking [286], and this was not a unique case [287]!

The damage in figure 21 is hemorrhagic but otherwise strikingly similar to the pattern of damage caused by asphyxia at birth. Wernicke's encephalopathy has been observed in animals inadvertently maintained on diets lacking thiamine [288, 290] and in experimental thiamine deficiency [289, 291]. Damage to the inferior colliculi was found most prominent in animals, in contrast to more severe involvement of the mammillary bodies in human cases of chronic alcohol use.

Damage caused by long-term escalating abuse of alcohol is serious, but not as catastrophic as sudden total deprivation of the essential aerobic coenzyme thiamine. Protective mechanisms spare the inferior colliculus and leave the mammillary bodies more vulnerable to repeated damage during the lifespan of a person addicted to alcohol.
[Top]

33 - Brain-Gut Relationship
Chronic alcohol use damages the esophagus, intestinal tract, and liver. Liver encephalopathy or cirrhosis is the terminal stage of alcoholism. Alcohol in excess directly damages these organs. But considering the speed with which mental changes and inebriation occur following use of alcohol, it seems clear that alcohol's first target is the brain.

Toxic impairment of the brain involves brainstem centers, and involuntary functions such as intestinal peristalsis depend upon intact brainstem circuits that control the autonomic nervous system [292]. Thus the gut is as dependent upon integrity of brainstem function as the brain is on absorption of essential nutrients like thiamine.

Myers referred to the brainstem pattern of damage by asphyxia at birth as a "monotonous rank-order of brainstem nuclei." This pattern of damage did not produce cerebral palsy, but it is a pattern of damage no less serious. Impaired autonomic function caused by brainstem damage should be investigated as possibly contributive to the gastrointestinal problems of some children with autism.

Korsakoff (1889) observed the signs of neurological impairments (oculomotor, ataxia, and mental confusion) described by Wernicke not only in cases of chronic alcoholism, but also during the course of infections and cancerous cachexia (or wasting which may also have involved thiamine deficiency) [293]. Korsakoff provided description of 14 cases of non-alcoholic origin. Among these were post-partum illnesses, typhoid, tuberculosis, tapeworm, diabetes, pneumonia, jaundice, and intestinal disorders. In addition to early acute symptoms similar to those reported by Wernicke, Korsakoff is best remembered for describing the long-term course leading to memory impairment.

Neubuerger (1937) discussed the possibility that "auto-toxic" substances could be produced in senile and diseased organs, and that these auto-toxins were the reason Wernicke's encephalopathy developed in the brain [294]. The brain-gut relationship continues to be described [295-298].

Morley (2003) has noted that immediate clamping of the umbilical cord at birth can leave the infant in a state of hypovolemic shock, which in turn triggers generalized vasoconstriction that shifts blood flow from less vital organs to the heart and brain [299]. Hankins et al. (2002) provided evidence that asphyxia from placental abruption or umbilical cord prolapse during labor results in injury to the liver, kidneys, and heart as well as the brain [300]. Autism is associated with complications at birth, and some children with autism may then begin life with multiple organ injury as well as damage of brainstem autonomic centers.
[Top]

35 – Autism and Complications at Birth

36 - Umbilical Cord Clamping

SUMMARIES

Main Points:

1. The brainstem auditory pathway is susceptible to damage by all of the known medical conditions associated with autism, because it is the most metabolically active system of the brain.
   
   
2. Protective mechanisms prevent auditory system damage except in catastrophic circumstances like circulatory arrest, suffocation, or poisons that disrupt aerobic metabolism.
   
   
3. Partial oxygen insufficiency during late gestation or birth is one cause of cerebral palsy, whose victims in some cases learn language on time and demonstrate good to excellent cognitive and social skills.
   
   
4. Total oxygen deprivation during late gestation or birth selectively damages the inferior colliculi in the midbrain auditory pathway, and is proposed here as a possible cause of the language and auditory attention problems of children with autism.
   
   
5. Damage to the inferior colliculi has been noted in previously normal adults who have lost the ability to understand spoken language [Meyer et al. 1996, Johkura et al. 1998, Masuda et al. 2000].
   
   
6. Verbal auditory agnosia (VAA), an inability to distinguish syllable and word boundaries in rapid streams of speech, has been identified as an underlying problem in some children with autism [Rapin 1997].
   
   
7. Inhibitory neurons in the inferior colliculus appear to stop transmission of ongoing sounds, providing a means for detecting sound onset which may also be involved in detection of syllable and word boundaries.
   
   
8. Difficulty learning a foreign language, without accent, past the first decade of life may indicate that verbal auditory agnosia (VAA) is a normal affliction of aging, and most severe in the elderly who lose the ability to engage in conversation in noisy environments.
   
   
9. Most young children learn language by ear and in a foreign language environment as easily learn a second language.
   
   
10. Abnormalities of the amygdala, hippocampus, mammillary bodies, cerebellum, and inferior olives have been observed in the brain of individuals with autism; all are part of a subcortical pattern of damage caused by asphyxia and toxic substances.
    
    
11. Protective mechanisms may prevent visible damage in the inferior colliculi and allow more noticeable abnormalities in the rank-order of subcortical nuclei that are somewhat less metabolically active.
    
    
12. The subcortical pattern of damage may imply impairment of function within the inferior colliculi even in the absence of visible involvement.
    
    
13. Progressive involvement of the cerebral cortex was observed in monkeys subjected to asphyxia at birth, and indicates a mechanism that could explain failure of frontal lobe maturation in children with autism.
    
    
14. Complications at birth are recognized as common in children with autism but their possible significance underestimated.
    
    
15. Immediate clamping of the umbilical cord has become a standard practice within the past 20 years, but an infant who does not breathe right away will suffer at least a brief but potentially catastrophic period of oxygen deprivation.
    
    
16. In research on asphyxia at birth in monkeys, asphyxia was produced by preventing breathing and clamping the umbilical cord; visible damage of the inferior colliculi occurred in some monkeys subjected to as little as six minutes of asphyxia.
    
    
17. Immediate clamping of the umbilical cord ought to be considered as one possible reason for the increased incidence (or prevalence) of autism evident within the last two decades.