Shaken baby syndromeBMJ 2004; 328 doi: https://doi.org/10.1136/bmj.328.7442.720 (Published 25 March 2004) Cite this as: BMJ 2004;328:720
All rapid responses
The pathologist challenging shaken baby syndrome
Irene Scheimberg states the below:
"I'm exploring all sorts of theories. My colleague Marta Cohen from
Sheffield Children's Hospital and I have just published a paper with
observations of our autopsy work on fetuses and babies over the last
couple of years. We selected 55 cases - 25 late third trimester fetuses
who died shortly before delivery and 30 newborns - who had haemorrhage
within the membrane that covers and separates the two halves of the brain,
and compared this with the level of brain hypoxia, or oxygen deficiency.
We knew that none of these cases could possibly be inflicted trauma. We
found that all those with severe brain hypoxia and half of those with
moderate brain hypoxia had SDH. This is the same type of SDH that some
people describe as specifically indicative of shaken baby syndrome. A
similar pattern of haemorrhages has been described in the retinas of
newborn babies dying of natural causes. We think that in these cases the
haemorrhaging is caused by the hypoxia." 
Asking for Risdon et al to apologise and accept they were wrong is
probably asking too much, I hope that when they read the latest research
they will, however, take note of it when deciding on causes of death. Ms
Geddes wasn't so far off the mark was she?
No doubt somebody is now going to claim that pregnant women were
undertaking activities that bounced the unborn child around in order to
harm them pre-birth, I'm not kidding either!
It's a shame that neither the BMJ Group or The Lancet chose to
publish these very important findings.
At the moment babies suspected of having been shaken are administered
powerful and potentially devastating anticonvulsants in large quantities
upon admission to hospital, the doctors treating what they believe to be
SBS - problem is, if it was a hypoxic event caused by any one of a number
of conditions known to cause such an event, then the treatment itself
I am and have been involved as part of a team researching medical records in alleged cases of SBS in both family and criminal cases
Competing interests: No competing interests
Whilst Drs. Brian Harding, R Anthony Risdon, and Henry F Krous were
correct in stating that other signs should be observed in Dx of SBS, they
kept returning to the "triad" as the core of the diagnosis. Each component
of said triad has many possible causes. How can such a diagnosis be made
based upon said triad except through dogma?
The Geddes et al. study does show pictures of non-traumatic, massive SDH,
it refers to macroscopic SDH's. It does state that several of the 50 non-
traumatic cases studies had RH's.
No human can inflict the G-forces by shaking, that is exerted on a
passenger in a 40mph+ motor vehicle collision. In such an instance, adults
and children whose heads do not impact the inside of the car generally
have cervical injury. They just don't present with SDH, RH and DAI, unless
they have head impact.
I am interested to know why given the questionable critera currently used
for SBS Dx, none of the cases using those criteria have been witnessed,
and why they are in children only under 12 months as stated by Harding et
al.. What changes when a child is over 12 months old? Do perpetrators
suddenly choose not to shake one year olds?
Robert Murdoch AP
Competing interests: No competing interests
Regarding the question of the Shaken Baby Syndrome (SBS) Drs Harding,
Risdon and Krous state, “a need exists for an impartial and intelligent
assessment, but how may this be achieved in practice?”
I suggest that since the role of vaccines in SBS has been questioned
[1,2,3,4,5] the history should start with the question, “when was the
child last vaccinated?”
Of the 20 reports I have on the subject:
9 were vaccinated between 0 – 21 days prior to the onset of their
signs and symptoms. Apnoea, Subdural and Retinal haemorrhage, Cutaneous
bruising and “fractures”.
8 had documented evidence of abnormal Haematological or Liver
Function Tests (LFTs) and Nutritional defects.
3 were incompletely documented.
This is proof that the administration of vaccines is one cause,
repeat one cause, of the so-called Shaken Baby Syndrome since it
demonstrates that SBS was confined to the period LESS THAN 21 DAYS AFTER
THE VACCINE WAS ADMINISTERED (the period in which hypersensitivity
reactions are expected to occur).
With the enormous number of persons accused, and for that matter
imprisoned, it should not be an impossible task for those believing in the
existence of SBS to negate this deduction by presenting documented
evidence of the condition occurring AFTER the 21 day post vaccine period,
having excluded haemorrhagic, liver and nutritional disorders.
The Medical Profession should be looking for evidence of deficiency
of Vitamin C induced by Immunization  and for Immune Complexes and
other tell-tale signs of Hypersensitivity to Vaccines.
It is time to change the name from ‘Shaken Baby Syndrome’ to ‘Adverse
Vaccine Reaction’ in those cases in which the condition follows the
administration of a vaccine within 21 days
Iatrogenic Misadventure would be another label worth considering for
the vaccine induced cases. The haematological and nutritional cases should
be appropriately designated. Vitamin K Deficiency Bleeding (VKDB) or
Barlow’s Disease would probably cover most of them.
Finally, the safety of vaccines can be assured by doing what
Kalokerinos did in the Australian Outback - administer Vitamin C before
immunizing the infant and ensure the adequacy of Vitamin K..
1.Kalokerinos A Every Second Child. With a Foreword by Linus Pauling.
Keats Publishing. Inc New Canaan, Connecticut. 1981.
2.Goodwin J Was It Murder Or A Bad Vaccine. Redbook Magazine
September 2000:158-175 Statement of Amici Curiae in support of Alan
3.Scheibner V. Shaken Baby Syndrome. The Vaccination Link. Nexus Aug-
4.Clemetson CAB. Vaccinations, Inoculations and Ascorbic Acid. The
Journal of Orthomolecular Medicine 1999 vol 14 p 137 – 142
5.Clemetson CAB Shaken Baby Syndrome or Scurvy.
I have given evidence for the accused
Competing interests: No competing interests
According to Harding, Risdon and Krous, the "triad" is defined, in
their opening statement as, "acute encephalopathy with subdural and
retinal haemorrhages, occurring in a context of inappropriate or
inconsistent history and commonly accompanied by other apparently
inflicted injuries." This definition of the triad is loaded with
inconsistencies. All that one needs to do is review the literature.
First of all, Caffey's original work, that is frequently cited as the
keystone of the SBS theory...yes, it is still a theory, is full of
inconsistencies. Caffey first proposes that the trio of symptoms/signs is
subdural hemorrhage (SDH), retinal hemorrhage (RH) and fractures of the
long bones or ribs. Then the trio of signs changes to SDH, RH and no
external signs of abuse. (1,2) Then in 1974, the triad is bilateral SDH,
bilateral RH no external signs of abuse and evidence of periosteal
traction of the long bones. (3) Caffey does mention that instances of
picking an infant up by an arm or leg, “swinging” or tossing the infant
through the air, during play, by the extremeties (a common mode of play
for parents/care takers-emphasis mine) can cause the bone lesions or
“traumatic involucra”. Infants shoulders have been known to dislocate
from picking them up by an arm, so to assume that it requires violent
shaking or slamming...to stretch the periosteum is beyond reason.
Of interest, in 1974, Caffey's paper includes the description of
necropsies on Baby's "H" and "K". Baby H had “skin normal, thymus large
microscopic focal hemorrhages in the myocardium: pinkish cellular exudates
in the pulmonary alveoli: small subcapsular laceration of the liver filled
with fresh blood, liver capsule intact.” Caffey seems to neglect the
information pointing to a differential diagnosis and focuses only on the
brain and eye hemorrhages. Does he suppose the bleeding in the myocardium
(heart) and pinkish cellular exudates in the alveoli (lung infection?) are
irrelevant? It would be far fetched to attribute the myocardial
hemorrhages to acceleration/deceleration, let alone the cellular exudates
in the alveoli. Perhaps Krous and Harding can tell us how this fits into
Baby K, as described by Caffey, basically had bleeding in the brain
and spinal cord. Baby K’s eyes were not examined. There was no evidence
of fractures. However, Baby K had “several small foci of atelectasis in
the lungs” with signs of generalized cyanosis. “Infant turned grayish and
died 2 hours after admission.” Why was there no consideration of the lung
findings, possible pneumonia or obstruction to explain the atelectasis and
generalized cyanosis (a sign of severe apnea or hypoxia)? Again, it seems
rather important differential diagnosis information was neglected and the
focus was solely on bleeding in the brain. The recent work by Geddes et
al. accounts for the brain hemorrhages in light of severe hypoxia...
Also, this infant only had clear evidence of one sign of the triad (SDH),
but shaking was still assumed to be the diagnosis.
The experimental evidence Caffey was greatly lacking came from
Ommaya, and Ommaya in 2002, states that Caffey misinterpreted his results
and their applicability to his theory. (9)
In 1984, a study by Ludwig and Warman add that “the physical findings
of tense or bulging fontanelle, head circumference greater than 90th
percentile for age, and retinal hemorrhage strengthen the diagnosis”. (4)
Ludwig and Wartman seem to be saying that a CT scan with bleeding in or
upon the brain is the sole criteria for diagnosis, but that retinal
hemorrhages and a swollen brain (bulging fontanelle) help strengthen the
diagnosis. Thus, SDH = SBS, but hydrocephalus (bulging fontanelle and
head circumference greater than 90th percentile) and RH help confirm the
diagnosis. They state this without any consideration to the meaning of a
head circumference greater than the 90th percentile and all of its
implications to other differential diagnosis. They do not discuss long
bone or rib lesions/fractures.
Then in 1987, Duhaime et al. perform a biomechanical study that
concludes shaking cannot generate enough force to meet the injury
threshold to cause subdural hemorrhage and traumatic axonal injury. (5)
Hence, the derivation of shaken baby/impact syndrome. To continually
assume shaking has anything to do with the actual injury is difficult to
comprehend. It’s like saying that if you jump up and down 9 times
(shaking), and on the 10th jump you land on someone’s foot (impact) which
results in you breaking your ankle; that the 9 prior jumps played a role
in the ankle fracture. Had the other persons foot not come into play,
numerous more jumps could have occurred without injury. The injury
occurred at the moment when another’s foot was landed on, and had nothing
to do with prior, non-injury producing, stresses. The ankle fracture
would have likely occurred had the person landed on someone else’s foot on
the first jump.
Then in 1989, Zimmerman and Bruce follow this study up by stating
that shaking alone cannot cause such injuries. They also state that “with
pure shaking impact injury, evidence of other trauma is rarely found”. (6)
Not according to Duhaime and the current letter by Krous and Harding who
state extracranial signs of trauma are common. Which is it? If the
proponents of SBS cannot even keep it straight, how is anyone else
supposed to? Zimmerman and Bruce were not the only ones to say the other
evidence of trauma is rarely found, just as Duhaime and Krous...are not
the only ones to say it is commonly found.
Then, in 2003, a study by Prange et al. concludes that “there are no
data showing that the maximum change in angular velocity and peak angular
acceleration of the head experienced during shaking and inflicted impact
against unencased foam is sufficient to cause SDHs or primary TAIs in an
infant”. (7) This REPEATS the earlier data that shaking alone cannot cause
subdural hemorrhage or traumatic axonal injury (this would seem to be
supported by the hypoxic injury found by Geddes that the axonal injury is
not trauma related). In addition, to result in SDH, it seems to require a
hard impacting surface. So, why do many alleged SBS cases lack evidence
of impact (the differential diagnoses is a completely different paper)?
Wouldn't an impact from a multi-story fall or high speed car crash (that's
what the "experts" tell us causes SDH or RH...that or shaking) leave
significant signs of impact?
To conclude, as you can see, the SBS "triad" changes with the
seasons...the seasons of data. The child abuse "experts" clothe the
"triad" depending upon the weather/season. How does Harding and Krous
propose the incongruent history given by caretakers be confirmed? That
all caretakers of children with a SDH get a polygraph? Maybe the SBS docs
should be interrogators instead of physicians? Who's story wouldn't
change after ten interrogations, threats, promises of leniency if you
confess... Furthermore, I speculate that many investigators routinely
change statements around? We know the "experts" won't accept any history
beyond a multi-story fall or high speed MVA to account for SDH or RH
despite all the evidence to the contrary...(i.e. Lantz's 14 month old
television tragedy, Geddes work, Plunkett's eye witnessed "trivial" falls,
Duhaime's shaking work...and that is just the tip of the iceberg). I
would assume that Harding, Risdon and Kraus added trauma to the neck and
spinal cord as a result of the work by Geddes et al, and Ommaya et al.
(8,9) Yet, both SDH and RH also occur during birth pretty frequently and
now we see them with a television toppling over onto one's head.
To further conclude, This is why I propose a new mechanism or feature
for the, what is now known as, shaken baby/shaken impact syndrome. I
propose that hitherto and henceforth, the syndrome be called
shaken/slammed/SQUASHED baby syndrome. Squashing coming from the
biomechanics of birth, TV compressing one's head... Call it squeezed,
compressed, squished...whatever you like. That's it folks SBS is now
characterized as first shaking, then slamming and if that doesn't work,
then the head is squeezed between two hands like it's in a vice. Don't
you see it, it's clear, definable, pathognomonic and beyond a reasonable
doubt or medical certainty??? But then again, the world might actually be
flat afterall, too?
L. Travis Haws
1)Caffey J. On the Theory and Practice of Shaking Infants. American
Journal of Diseases in Childhood 1972; 124:161-9.
2)Caffey J. The parent-infant traumatic stress syndrome: (Caffey-Kempe
Syndrome), (Battered Babe Syndrome). Amer J Radiol 1972; 114:218-29.
3)Caffey J. The whiplash shaken infant syndrome: manual shaking by the
extremities with whiplash-induced intracranial and intraocular bleedings,
linked with residual permanent brain damage and mental retardation. Pediat
4)Ludwig S, Warman M, Shaken Baby Syndrome: A Review of 20 Cases. Annals
of Emergency Medicine Feb. 1984; 104-7.
5)Duhaime A. et al., The Shaken baby syndrome. A clinical, pathological,
and biomechanical study. J Neurosurg 1987; 66:409-415.
6)Bruce DA, Zimmerman RA. Shaken impact syndrome. Pediatric Annals 1989;
7)Prange M. et al. Anthropomorphic simulations of falls, shakes, and
inflicted impacts in infants. Journal of Neurosurgery 2003 99: 143-150.
8)Geddes J. F. et al., Dural haemorrhage in non-traumatic infant deaths:
does it explain the bleeding in ‘shaken baby syndrome’? Neuropathology and
Applied Neurobiology 2003; 29; 14-22.
9)Ommaya A., Goldsmith W., Thibault L., Biomechanics and neuropathology of
adult and paediatric head injury. British Journal of Neurosurgery 2002
Know The Falsely Accused
Competing interests: No competing interests
I am a research attorney who represents a man who was falsely accused
of murder in 1983. I have been researching shaken baby and fatal head
injury articles for the past four years. I feel I must respond to the
Harding, Risdon, Krous editorial that stands for the proposition that
Shaken Baby Syndrome is a valid scientific medical diagnosis and cites to
Case ME, Graham, MA, Handy TC, Jentzen JM, Monteleone JA, Position paper
on fatal abusive head injuries in infants and young children. (Am J
Forensic Med Pathol 2001:22: 12-122.)
Harding, et al. do not disclose that the Case, et al. position paper
did NOT pass peer review. According to the AJFMP Editor's note on page
1056: "The manuscript was reviewed by three reviewers on the Board of
Editors of the American Journal of Forensic Medicine and Pathology. They
believed that while it was worthy of publication, it should not be
published as a position paper because of the controversial nature of the
subject." The NAME (National Association of Medical Examiners) Board
nonetheless voted to publish the paper. Was there a conflict of interest?
While the Am J Forens Med Pathol does not have a written conflict-of-
interest statement, the American Journal of Forensic Science, a similar
publication although with a broader audience, does. According to the
American Journal of Forensic Science website, (adapted from the
International Committee of Medical Journal Editors Statement on Conflict
of Interest): "A conflict of interest for a given manuscript exists when a
participant in the peer-review and publication process--author, reviewer,
or editor--has ties to activities that could inappropriately influence
his/her judgment, whether or not judgment is in fact affected....Public
trust in the peer-review process and the credibility of published work
depend in part on how well conflict of interest is handled during writing,
peer review, and editorial decision making. Bias can often be identified
and eliminated by careful attention to the scientific methods and
conclusions of the work."
At least two of the "paper" authors, Case and Graham, were on the
NAME Board of Directors at the time the paper was published in the journal
despite being rejected through peer review. Coincidentally, all of the
paper's authors except Case are members of the Helfer Society. (See
www.helfersociety.org/Members.htm) Dr. Krous is also currently listed as a
"Helfer" member, as are many of the SBS proponent authors, such as Drs.
Alexander, Chadwick, Christian, Dias, Feldman, Hymel, Jenny, Kleinman, and
Levin. The U.K’s Sir Roy Meadow is also an honorary lifetime Helfer
Many of the Helfer members speak at the various National Center on
Shaken Baby conferences and other SBS oriented conferences. In advocating
the existence and science of the syndrome, they routinely cite to another
Helfer’s article. Is there a competing interest here?
The time has come to critically evaluate the biomechanics of head
injury in children instead of blindly adopting articles written by the
child protection oriented physicians. The medical, scientific and legal
communities must closely examine whether there is a competing interest and
whether these articles involve qualitative research and studies.
Tracy Emblem, Esq.
Competing interests: No competing interests
First I want to thank the BMJ for not only making its content free to
all (for the time being) but for allowing us the opportunity to voice our
In response to the Harding, Risdon and Krous piece concerning the
disgnosis of shaken baby syndrome I am much concerned that they cannot see
the "whole picture."
"The basic triad should have all the necessary features for confident
diagnosis and the conclusion that undue force has been applied."
By citing opinion pieces and reports which would clearly fall in the
time frame of Mark Donahoe's recent review of the literature, it is clear
that they refuse to admit the basis of their conclusions is weak and
getting weaker everyday.
Even in the light of the revelation that retinal findings that were
previously attributed to violent shaking are not necessarily caused by
such they continue to spout the "typical" war cry of the "classic" shaking
baby syndrome victim.
The parallels drawn in conclusions of the study by the Wake Forest
University team and of Mark Donahoe on the available literature should be
enough to make any medical servant take pause. The recent reports by
Geddes et al, Plunkett and even Duhaime (I did note the lack of inclusion
of her lastest report published in 2003) lead the reader to agree that
indeed we don't know what is happening. Can you truly claim that an
unwitnessed accidental fall is sbs in light of the findings of the past
several years? We have seen the proof that falls can generate sdh and rh
and glocal brain swelling.
We have evidence that hypoxic injuries can also cause similar
bleeding in infants who have not suffered trauma. Clearly the evidence is
indicating that there may indeed be a likelihood that trauma may not be a
factor in infantile subdural hemorrhage and that the retinal findings may
most likely be related to the intercranial pathology not a sign of
I just read a report "Frequency and natural history of subdural
haemorrhages in babies and relation to obstetric factors" that showed a
shocking number of infants with asymtomatic sdh following the birth
process. Some have used this report to try to dispel birth injury claiming
that since the sdh in these nine infants resolved by four weeks then ALL
sdh seen after four weeks cannot be attributed to birth injury. I read the
report and saw nine very fortunate infants with diagnosis made which
clearly affect their everyday handling and care.
Continuing to say that rotational/deceleration causes these findings
without having the evidence to back it up is not going to make this an
accurate diagnosis. In light of the recent evidence it has become clear
that nonviolent etiology may be indicated and certainly needs to be
Competing interests: No competing interests