An Introduction to the Scientific Controversy over Shaken Baby Syndrome

The scientific controversy over the so-called Shaken Baby Syndrome (SBS) (also known, with a few key differences indicated later, as Abusive Head Trauma, AHT) is a rich and highly complex multifaceted topic.

It involves a wide range of academic disciplines: pediatrics, neonatology, obstetrics, neurology, neuropathology, radiology, hematology, evidence-based medicine, biomechanics, statistics, epidemiology, psychology, among others. There are also ramifications in epistemology, medical ethics, criminal law, sociology...

As such, the mass of data and information collected in the past half-century is phenomenal. There has been hundreds of popular press articles and documentaries in the past decades, while the academic publications number in the thousands.

Investigating this hotly-debated subject is a challenging and even daunting endeavor.

Yet, contrary to many other contemporary academic topics, it remains practically feasible for a single, or a small group of, highly-motivated individual(s) to read a large part of the literature available and gain a deep comprehension of the relevant issues.

This task is not even reserved to doctors and researchers. With enough time and energy, anyone with acceptable academic reading skills and a minimal high-school level scientific background can form a robust, evidence-based, objective opinion by carefully learning about both sides of the controversy. That includes lawers, judges, law enforcement officers, social workers, and other professionals involved in these cases.

This post provides a few pointers to anyone interested in understanding the most pressing issues around SBS. It is by no means exhaustive. It suggests a bare minimum of essential resources one should look at before delving into more specific aspects of the controversy.


tl;dr : Summary of the current state of scientific knowledge at the time of writing (November 2022):

  • Inflicted head trauma may cause a constellation of extracranial and cranial injuries in infants, including intracranial and ocular hemorrhage.
  • Accidental and non-accidental impact to the head may cause subdural and retinal hemorrhage in infants.
  • The presence of multiple traumatic injuries on various part of an infant's body raises suspicions of abuse.
  • Whether shaking alone, without impact, may cause rupture of bridging veins, subdural hemorrhage, and retinal hemorrhage in healthy infants, without causing additional traumatic injuries to the neck and other parts of the body, remains controversial.
  • Whether the sole observation of subdural hemorrhage, bridging vein rupture, and retinal hemorrhage in infants, without additional pointers to abuse, can lead to a near-certain medical determination of abusive head trauma, remains controversial.
  • There is general agreement that brain injuries in SBS/AHT are hypoxic in nature and not traumatic.
  • The scientific value of confessions of shaking obtained after medical determinations of SBS/AHT is debated, and whether they offer a posteriori validation of these determinations remains controversial.
  • There is general scientific agreement that videotaped and witnessed cases of shaking are extremely rare, and that most of these rare cases do not appear to result in intracranial and retinal hemorrhage. The scientific and forensic significance of this observation is debated.
  • There is general scientific agreement that biomechanical and animal models have so far failed to reproduce the cardinal findings associated to SBS/AHT, but whether this weakens the SBS/AHT hypotheses, or whether it reflects failure of the models to accurately reproduce pediatric traumatic brain injuries, remains controversial.
  • There is general scientific agreement that benign external hydrocephalus may sometimes be associated with isolated subdural and retinal hemorrhage in infants, without major trauma.
  • There is general agreement that short falls may rarely cause severe intracranial injuries, including subdural and retinal hemorrhage, and/or death, although the exact frequency of occurrence remains debated.
  • There is universal agreement that shaking and other forms of inflicted head trauma and child abuse are dangerous and justify criminal prosecutions. Public policies focusing on prevention are warranted.
Post updates
2022-11-07added citations
2022-10-17minor updates
2022-10-05added summary
2022-10-03highlighted references
2022-09-30initial version

Table of contents


Before presenting some of the most important references on the subject, we will indicate the caution required when investigating the subject, and we will try to clarify the definitions and the most contentious points.

Caution to reader

It is crucially important to understand that SBS/AHT is not an academic subject like others, for reasons listed here:

  1. SBS/AHT is not just an abstract controversy. It has major, direct judicial implications in courts all over the world on a daily basis (removal of children from their homes, criminal convictions of parents and caregivers). The judicial decisions pronounced on the basis of medical expert opinion on SBS/AHT have immediate consequences for the affected children and their families.

  2. SBS/AHT is related to the delicate topics of child abuse and child protection. It specifically involves babies, including severely disabled or deceased infants. These concepts naturally elicit strong emotions in most individuals. This may sometimes hamper objective thinking in otherwise rationally-minded professionals such as doctors, scientists, judges, and lawyers.

  3. This academic subject has major societal and political ramifications. Child protection is an important aspect of most Western public policies. Disclosing or recognizing possible systemic failures of governmental agencies and processes bears some political risk.

  4. There is a significant amount of confusion and even obfuscation in the literature regarding terminology and definitions. The meaning of commonly-used terms has slowly, and often implicitly, evolved over the course of decades. Discrepancies only appear when confronting literature decades apart.

  5. Terminology is often vague, unclear, and imprecise. Descriptions are often qualitative rather than quantitative. For example, the degree of force associated to an act of shaking is not described in terms of velocity or acceleration, but as an act "so violent that individuals observing it would recognize it as dangerous and likely to kill the child" (AAP 2001). These descriptions render scientific investigations difficult while making it clear they are designed for legal purposes.

  6. The degree of polarization in this area is extreme. As a first approximation, one may consider a currently dominant view, supported by what we may call "SBS/AHT adherents", and a different view supported by "agnostics". Even the very existence of a legitimate controversy is denied by most adherents. We will discuss these terms later in this post.

  7. The adherent side is presented as having near-universal institutional support. It is fundamentally important that, for the sake of making a robust, evidence-based scientific opinion, one must make a sustained effort to move beyond any argument from authority. The reader should not substitute collective certainty with scientific evidence. Although in general, institutional support may be expected to follow scientific evidence, this is not strictly guaranteed, particularly when there are major non-scientific (political, financial, societal...) ramifications. Thus, one should not rely on one's natural confidence in scientific authority; rather, the reader should keep an open mind and exercise critical thinking at all times, and never hesitate to question or double-check any explicit or implicit assumption.

  8. As a corollary to this extremely polarized dispute, the field is replete with violent personal attacks, insults, ad-hominem arguments, straw-man arguments, misstatements, and misrepresentations. The reader needs to keep a calm, objective, rational mind and concentrate on nothing other than robust methodology and sound scientific evidence, leaving all emotion aside.

  9. A superficial read of the SBS/AHT academic literature, especially reviews, is far from sufficient to gain a deep understanding of the issues at play. One has to comprehend their historical context and evolution over periods spanning many decades. It is often necessary to look up and read cited references repeatedly.

  10. One must develop the unnatural habit of checking every cited and seemingly literature-supported item of information. It has been shown on multiple occasions that certain assertions cannot be accepted at face value, even when references are proposed. Examples may be given later.

  11. Contrary to the philosophy of the open science movement, anonymized data on which medical studies are based is almost never available. This is partly explained by obvious issues related to the medical and judicial confidentiality of child abuse cases. For example, when a clinical study includes cases that received a medical determination of child abuse, the exact medical information used by researchers to make these determinations is not accessible, or in an extremely limited form (as a table or a graph in the publication). When cases are selected based on the existence of confessions obtained during police interrogation, the context and specifics of the interrogation are not available. Since independent verification is impossible, the scientific value of most studies is limited by the opacity of the data they rely on. In this context, testimonies of individuals relating personal or professional experience of these cases may sometimes bring valuable insight.

  12. When writing about this topic, it is often necessary to make obvious statements explicit, such as recalling that shaking a baby must be strictly avoided, that severe injuries are to be expected after inflicted head trauma in children, and that child abusers must be criminally prosecuted. The reason for stating the obvious is that SBS/AHT agnostics are often accused, without any supporting evidence, of "denying" the existence of child abuse, or pretending that "shaking is safe". In reality, the controversy pertains uniquely to the reliability with which child abuse can be inferred with a high degree of certainty based on a few medical findings alone.

"Adherents" vs "agnostics"

We will refer to both sides of the controversy as "SBS/AHT adherents" and "agnostics". This is not a perfect denomination, and in any case any choice of denomination can be considered biased by one side or another. We intend these terms to be understood precisely according to their definitions of the Collins dictionary:

  • Adherent: "someone who holds a particular belief or supports a particular person or group"

  • Agnostic: "a person who claims, with respect to any particular question, that the answer cannot be known with certainty"

Adherents generally believe that SBS/AHT is a valid, non-controversial medical diagnosis that has an extremely low rate of false positives (Narang 2016).

On the other side, whereas a very small number of authors may deny the "existence" of SBS/AHT, most merely point to a lack of scientific reliability and certainty about these medical determinations of abuse and, therefore, a possibly high number of wrongful convictions and miscarriages of justice. Here is a citation from some of these authors:

"Today, we are still seeking answers to the questions that we have been asking for 40 years or longer—questions such as, why do some infants or toddlers suddenly collapse or die? Why do some of these children have subdural hemorrhages while others do not? What does the presence of the triad (or some elements of the triad) tell us about the cause of the collapse or death? And are there any findings that can accurately distinguish between accidents, abuse and natural causes? For decades, we thought we had answers to some of these questions: we thought that the presence of the triad, or some of its elements, proved that the child had been shaken. Today, the correct answer to these questions is, 'we don't know.' And, until we do know, we are, in Dr. Duhaime's words, simply 'shooting in the dark.'" (Findley 2011, pp. 261-262, emphasis ours)

A tentative definition

A required first step when investigating such a subject is to be clear on definitions. The SBS/AHT controversy is so intense that even the most basic definitions are vigorously debated. The term "Shaken Baby Syndrome" has actually several totally different meanings, which is a major reason for the high degree of confusion and obfuscation surrounding this issue.

On the one hand, Shaken Baby Syndrome describes a child abuse act where an adult exasperated by the uncontrollable crying of a young baby holds the child under the arms and vigorously shakes him or her back and forth. The brutal and repeated angular movement sustained by the baby's head during this act, amplified by the weakness of neck muscles and relatively large size of the infant head, may cause neurological injuries. Severe disability or death may ensue. These dramatic consequences on the child justify prevention measures targeting parents and caregivers, but also healthcare professionals who should be trained to recognize characteric signs in young victims.

On the other hand, Shaken Baby Syndrome also describes a set of related scientific hypotheses that we will formulate as follows (medical terms will be explained briefly below):

  1. Violent shaking causes rupture of bridging veins, intracranial hemorrhage, retinal hemorrhage, and brain injuries in infants. Other traumatic injuries to the head, neck, ribs, limbs, organs, and other parts of the body are possible, but absent in many cases.

  2. When a child is violently shaken, severe neurological symptoms appear almost immediately (the child instantly loses consciousness or will undergo a dramatic and sudden change of mental state).

  3. The only other possible causes of these medical findings are: a multistory fall, a high-speed motor vehicle accident, and extremely rare medical conditions such as glutaric aciduria, Menkes disease, some severe blood clotting disorders, and a few others. In other words, the medical findings mentioned above are thought to have a traumatic origin in nearly all cases. If no trauma is evoked by parents or caregivers, the only logical explanation is that they are lying.

Consequently, we propose the following tentative definition:

SBS Definition : Shaken Baby Syndrome refers to the idea that if an infant is found with a constellation of medical findings including one or several of: subdural hemorrhage (SDH), retinal hemorrhage (RH), and/or bridging vein thrombosis; if no history of a major trauma such as a multistory fall or a high-speed motor vehicle accident is given by the parents or caregivers; and if none of the few rare medical conditions mentioned above is found, then it can be concluded, by default but with near-absolute certainty, that the child has been violently shaken by the last person present with him or her when symptoms first appeared.

The first definition given above, related directly to an act of child abuse, is obviously non-controversial; all debate concerns only the second.

Surprisingly, a clear definition of SBS/AHT such as ours is rarely found in this explicit form in the medical literature, although it does appear in a few instances. It mostly emerges from the literature corpus as a de facto medical belief taught and shared by healthcare providers around the globe since the late 1970's and up to the 2010's. An abundant case law over the last five decades across many countries also shows that medical experts have long relied on this hypothesis in family and criminal courts.

Before we pursue, we must give some basic medical terminology.

Some medical terminology

  • The brain is surrounded by three meningeal membranes: the pia, the arachnoid, and the dura mater.
  • The space between the pia and the arachnoid is the subarachnoid space. It is filled with cerebrospinal fluid. The arachnoid and the dura mater are in continuity with one another, forming the subdural compartment which may contain fluid or blood.
  • Subdural hemorrhage, blood in the subdural compartment, is found in up to 46% of newborns (Rooks 2008). An intracranial hemorrhage occurs when blood is found in any of the tissues or membranes within the head.
  • Bridging veins drain the blood from the brain, crossing the subarachnoid space to enter sinuses within the dura.
  • Retinal hemorrhage occurs when blood is found in the retina, at the back of the eyes. It is found in about 20% of newborns (Callaway 2016).
Meninges of the central nervous parts. Credits: Wikimedia, SVG by Mysid, original by SEER Development Team, Jmarchn.

Legal implications

Our tentative definition of SBS has clear legal implications. Infantile subdural and retinal hemorrhage are considered to be so specific for violent child abuse that their sole presence in an infant, even with no sign of trauma on the body and no suspicion of abuse in the child's environment, constitutes an almost irrefutable proof that a criminal offence has occurred and, as such, justifies drastic child protection procedures, criminal prosecutions, and convictions.

In practice, medical determinations are almost always made similarly. Parents or caregivers bring a young child (most often aged less than 6 months, almost always less than 12) to the emergency room for subacute or acute symptoms such as seizures, fussiness, irritability, apnea, poor eating, vomiting, increasing head circumference... The child undergoes cranial CT or MRI and doctors find subdural collections and/or hemorrhage. Fundoscopy reveals retinal hemorrhage. The hospital team makes a medical determination of SBS/AHT and reports the family to authorities. Social and police investigations may lead to the removal of the child (and siblings, if any) from their familial environment, while parents or caregivers are prosecuted. At the trial, medical experts testify that their determination of child abuse is nearly certain.

We see here that the issue of SBS is really situated at the intersection of medicine, science, and law.

The degree of certainty with which medical experts determine the occurrence of abuse in these cases is important to the courts since, in common law jurisdictions, the legal standard of proof is "beyond a reasonable doubt". It is this certainty that can lead to criminal convictions, even when there is no other supporting evidence of child abuse beyond the medical determination of SBS.

🇫🇷  In legal systems based on civil law, such as France, the judges' and jurors' "intimate conviction" is the standard of proof (Esnard 2013). Here, too, the certainty with which highly-recognized and experienced judicial experts make a medical determination of child abuse has a direct impact on the "intimate conviction" of triers of facts. Presentations of slow-motion videos, or direct demonstrations, of an adult violently shaking a doll may also have a psychological impact on judges and jurors.

At the time of writing, the National Registry of Exonerations has identified 27 of exonerations of past convictions in SBS/AHT cases in the United States.

A contested definition

Norman Guthkelch, the British neurosurgeon who first evoked a possible causal link between shaking and infant subdural hemorrhage, has criticized the very term Shaken Baby Syndrome:

"In contrast, the appellation shaken baby syndrome (SBS) asserts a unique etiology (shaking). It also implies intent since it is difficult to 'accidentally' shake a baby. (...) Since subdural and retinal hemorrhages (with or without cerebral edema) may also be observed in accidental or natural settings, I suggest that the elements of the classic triad of retinal hemorrhage, subdural hemorrhage and cerebral edema would be better defined in terms of their medical features. Since subdural hemorrhages in infancy originate in the dura, perhaps "retino-dural hemorrhage of infancy" would be an acceptable name for the primary findings. (...) This would allow us to investigate causation without appearing to assume that we already know the answer." (Guthkelch 2012, p. 202, emphases ours)

This definition, criticized by agnostics, has also been largely abandoned by adherents.

In 2009, the American Academy of Pediatrics published a position paper recommending abandoning the use of Shaken Baby Syndrome, and replacing it with a newer and broader term, Abusive Head Trauma (AHT):

"Shaken baby syndrome is a subset of AHT. (...)

The term "shaken baby syndrome" has become recognized by the public (...) The American Academy of Pediatrics supports prevention efforts that reduce the frequency of AHT and recognizes the utility of maintaining the use of the term "shaken baby syndrome" for prevention efforts. Just as the public commonly uses the term "heart attack" and not "myocardial infarction," the term "shaken baby syndrome" has its place in the popular vernacular. However, for medical purposes, the American Academy of Pediatrics recommends adoption of the term "abusive head trauma" as the diagnosis used in the medical chart to describe the constellation of cerebral, spinal, and cranial injuries that result from inflicted head injury to infants and young children." (Christian 2009, p. 1410, emphases ours)

This definition shies away from considering isolated shaking as a specific abusive gesture. The newer term has a broader scope which encompasses all kinds of traumatic injuries inflicted to an infant's head, often involving direct impact to the head.

This change of definition only adds to the confusion, since it abandons the kernel of the issue: the link between shaking alone, without impact and purely intracranial findings. The association of non-accidental trauma involving impact to the head with severe injuries is not contested by anyone.

Agnostics also criticize the term Abusive Head Trauma on the basis that this term continues to conflate an intention (abuse) with a medical diagnosis. They contend that abuse is a legal determination, not a medical one (Findley 2019).

The "triad"

Alongside this shift of terminology, the SBS Definition indicated above was abandoned.

Historically, the term "triad" has been strongly associated with this definition.

The "triad" refers to three medical findings: subdural hemorrhage, retinal hemorrhage, and brain swelling. In the past, authors from both sides used this term to refer to the idea that the discovery of one or several of these medical findings alone, when unexplained by the story or by the discovery of the few rare medical causes, even without any traumatic injury to the skin, the bones, or the neck, suffices to "diagnose" child abuse with near-certainty.

We will abstain from using the term "triad", even if this term occurs frequently in the literature from both sides. It may lead to confusion as it is sometimes used when only one or two of its components are identified. Furthermore, these findings are likely to be pathophysiologically related, so their occurrence should not be considered as statistically-independent events.

Importantly, the very existence of the "triad", indicating a near-automatic determination of child abuse based on a few medical findings, is now denied by many adherents. Thus it is not just the scientific reliability of our tentative definition of SBS that is contested, but even whether this definition is used at all.

Indeed, adherents now claim that "diagnosing" SBS/AHT is a highly complex process that involves an entire multidisciplinary team, and that never relies on the sole observance of subdural and retinal hemorrhage:

"An abuse diagnosis relies upon careful review of all available data, often including data identified and assessed by a dedicated multidisciplinary team, a constellation of imaging findings in the brain, bones, neck, spine and abdomen, fundoscopic findings, interviews with caregivers, forensic data (including postmortem studies), the presence of additional or previous injuries to the child or siblings, the presence of other malicious injury (e.g., burns, bite marks) and exclusion of underlying diseases and accidental injury." (Saunders 2017, pp. 1386-1387)

Yet, even though the SBS Definition above has been officially renounced by several medical institutions, it continues to be used in the field by many medical professionals and judicial experts around the world. One of the reasons for this is the inertia of clinical practice and criminal justice.

🇫🇷  Curiously, France has adopted a peculiar stance in this international context by adopting national health guidelines in 2011 and 2017 that match almost perfectly with our tentative definition, but contradict the current international consensus (HAS 2017).

It should also be noted that, in the past ten years, the strong association of retinal hemorrhage with abuse has been slowly replaced by an association with severe retinal hemorrhage.

Citations from the medical literature, old and new

We give here a few relevant citations from the adherents' medical literature, illustrating a paradigm shift about the existence and validity of "triad"-based medical determinations of abuse between the late 1990's and the late 2010's. These citations are by no means exhaustive.

Older literature

"SBS usually produces a diagnostic triad of injuries that includes diffuse brain swelling, subdural hemorrhage, and retinal hemorrhages. This triad must be considered virtually pathognomonic of SBS in the absence of documented extraordinary blunt force such as an automobile accident." (Kirschner 1997, p. 272, emphases ours)

"The shaken baby syndrome (with or without evidence of impact) is now a well-characterized clinical and pathological entity with diagnostic features in severe cases virtually unique to this type of injury—[1] swelling of the brain (cerebral edema) secondary to severe brain injury, [2] bleeding within the head (subdural hemorrhage), and [3] bleeding in the interior linings of the eyes (retinal hemorrhages). Let those who would challenge these diagnostic features first do so in the peer-reviewed literature, before speculating on other causes in court." (Chadwick 1998, emphasis ours)

"The expert who acknowledges the classic findings of SBS include subdural hematoma, retinal hemorrhage and edema, but chooses to ignore this constellation in favor of an alternative hypothesis will appear foolish." (Holmgren 2001, p. 319, emphasis ours)

"Often referred to as the "triad", the consensus continues to be that a collection of (1) damage to the brain, evidenced by severe brain swelling and/or diffuse traumatic axonal injury; (2) bleeding under the membranes which cover the brain, usually subdural and/or subarachnoid bleeding; and, (3) bleeding in the layers of the retina, often accompanied by other ocular damage, when seen in young children or infants, is virtually diagnostic of severe, whiplash shaking of the head." (Parrish 2000, p. 1, emphases ours)

Newer literature

"Physicians experienced in the clinical evaluation of paediatric traumatic brain injury and AHT do not diagnose SBS by a ‘triad’. That is a ‘strawman’ created for legal purposes and legal recourse. (...) We know of no guideline or reputable paediatric treatise that recommends SBS/AHT be diagnosed by a ‘triad’." (Narang 2017, p. 1047-8, emphasis ours)

"AHT is not a diagnosis that is made, or excluded, based only on the presence of the three 'triad' factors, namely retinal haemorrhage, subdural haematoma and encephalopathy. Although these findings are commonly seen in AHT cases, to suggest that diagnostic processes simply rest on the presence or absence of the ‘triad', without regard to the specific features and clinical circumstances of the findings, is, at best, misleading. (...) We maintain that the term 'triad' carries no value for clinicians experienced in evaluating suspected AHT." (Lucas 2017, p. 1033, emphases ours)

"The diagnosis of abusive head trauma in children is not solely based on the findings of subdural hematoma, hypoxic–ischemic encephalopathy and retinal hemorrhage – it never has been." (Strouse 2018, p. 1045)

"Efforts to create doubt about AHT include the deliberate mischaracterization and replacement of the complex and multifaceted diagnostic process by a near-mechanical determination based on the "triad" – the findings of subdural hemorrhage, retinal hemorrhage and encephalopathy." (Choudhary 2018, p. 1050, emphasis ours)

Contested topics, open questions

We propose some open debated questions, with links to scientific references below.

Selected bibliography

In the remainder of this post, we will give references to learn more about the controversy.


  • We classified references between "SBS/AHT adherents" and "agnostics" when relevant. This distinction is not always clear and it is open to discussion.
  • Some references may appear multiple times in different categories.
  • References only available in French are indicated with a flag  🇫🇷.
  • Important references are highlighted in yellow. This selection is subjective and also open to discussion.
  • This bibliography will be completed over time. Contact me by email (see sidebar/footer on this page) to propose any corrections or additions.
  • Many articles are behind a paywall. Older articles may also be hard to find. Contact me if you need the PDF of some articles.


SBS/AHT adherents


Press investigations

In this section, we give references to press investigations specifically covering the scientific and legal controversy.

SBS/AHT adherents


The Waney Squier case

Waney Squier, a pediatric neuropathologist from Oxford University, UK, changed her mind about the scientific reliability of SBS/AHT in the early 2000's. She was struck off the General Medical Council (GMC) in March 2016, reinstated in appeal in November 2016, but she was not allowed to give expert evidence in court for three years.

Opinion pieces

SBS/AHT adherents


Films, documentaries, talks


Video interviews


Blog posts

PubPeer comments

Governmental reports

We give references to official position papers, or "consensus" papers, organized by countries rather than by side.

United States




Academic publications

In this section, we give some of the main references to medical, scientific, and legal academic publications, sorted by category and date of publication. There are about ten times more publications in the literature. We try to give references from both sides of the controversy, when this distinction makes sense.


The publications below are considered as the seminal works leading to the recognition of shaken baby syndrome. The two main figures are John Caffey (1895-1978), American pediatric radiologist, and Norman Guthkelch (1915-2016), British pediatric neurosurgeon.

Agnostics claim that initial works on SBS were based on anecdotal evidence and misunderstandings, notably misinterpretation of the study by Pakistani American neurosurgeon Ayub Ommaya in 1968 on whiplash injuries in monkeys. For them, SBS was applied in court before it was ever validated by robust scientific studies. In the last few years of his life, Norman Guthkelch has also criticized the way medical determinations of SBS were made.

Opiniated reviews

We give here references to reviews that may be found to present a bias toward one side of the controversy.

SBS/AHT adherents


Systematic reviews


We now move to more technical references.

Reading studies in neuropathology and forensic pathology ideally requires some minimal knowledge of the child's brain anatomy, physiological and pathological processes in tissues, cell biology, molecular biology, and immunostaining techniques.

By understanding the natural reactions of cells and tissues to pathological conditions such as mechanical trauma or hypoxia-anoxia (lack of oxygen), and by observing the tissues of deceased infants, pathologists can elaborate and verify hypotheses related to the causes of observed medical findings.

On the other hand, whereas radiology allows one to make observations of medical findings in the tissues of living patients, it is typically required to make a careful comparison with pathological findings in order to verify the observations and infer causative mechanisms, in particuler when evaluating the traumatic or non-traumatic nature of specific findings.

A significant aspect of the SBS/AHT controversy pertains to neuropathological discussions about the traumatic or non-traumatic nature of specific brain lesions observed in deceased infants, as this strongly determines whether child abuse will be ultimately retained as the cause of death or not.

Whereas traumatic brain injury seemed to be common in children who died with a medical determination of SBS/AHT, it has been shown in the last two decades that hypoxia is significantly more commonly observed than axonal damage, for reasons that remain unclear and debated.

Similar discussions also exist regarding bridging veins, subdural, and retinal hemorrhage.

Fig. 4 of Mack 2009. "a Left subdural compartment hemorrhage. a T1-W coronal MR image shows left subdural compartment hemorrhage. b Graphic representation of hemorrhage originating from the inner dural plexus (BV bridging vein, DP dural plexus)"

SBS/AHT adherents



We refer to the introduction of the Neuropathology section above.

Some authors remind that radiology alone may not yield a reliable medical determination of abuse:

"The medical and imaging evidence, particularly when there is only central nervous system injury, cannot reliably diagnose intentional injury. Only the child protection investigation may provide the basis for inflicted injury in the context of supportive medical, imaging, biomechanical, or pathological findings." (Barnes 2007, p. 53)

SBS/AHT adherents


Retinal hemorrhage

The pathophysiological mechanisms of retinal hemorrhage is incompletely understood. As a first approximation, SBS/AHT adherents believe they occur due to the vitreo-retinal traction during traumatic shaking, whereas agnostics believe they occur as a consequence of raised intracranial pressure and/or intracranial hemorrhage.

SBS/AHT adherents



Between a quarter and a half of all asymptomatic neonates present subdural and/or retinal hemorrhage in their first days of life. Although they typically cause no symptoms and resolve in a few days, they may persist and cause complications in a small proportion of cases.

An often-cited study is Rooks 2008 who found subdural hemorrhage within 72 hours of delivery in 46% of 101 asymptomatic neonates. The hemorrhage resolved spontaneously within three months in all cases, but one child had a new extraaxial collection on follow-up at 26 days which raised concern for abuse, which was ruled out after a full investigation. This case shows that birth intracranial hemorrhage may sometimes persist after several weeks and be misinterpreted as child abuse:

"Initially, this patient had bilateral posterior occipital SDHs, which were being followed for resolution. At 26 days of life, the patient returned for the follow-up MR imaging and was noted to have a 1-cm extraaxial left frontal collection that did not conform to CSF attenuation, consistent with a spontaneous SDH. The patient was admitted for full evaluation for nonaccidental injury to include skeletal survey, ophthalmologic examination, coagulation panel, metabolic studies, as well as social work enquiries. These investigations did not reveal any additional injuries or findings to support NAI [Nonaccidental Injury] as an etiology of the spontaneous frontal SDH (Fig 4). At a 5-month follow-up MR imaging, the left frontal SDH resolved; however, the subarachnoid space remained prominent in this patient. This finding suggests that though not typical in a neonate, prominent extra-axial space is a predisposing factor for SDHs as has been reported by other authors." (Rooks 2008, pp. 1087-1088 , emphases ours)

Fi. 4 of Rooks 2008. "Images obtained at 7 and 26 days postnatal age for follow-up of bilateral occipital SDH in a neonate with extra-axial collections. Axial T2, T1, gradient-refocused echo (GRE), and FLAIR images (left to right, top row) show CSF-intensity frontal subarachnoid collections that were present since birth. Also note a thin linear T1 hyperintense GRE hypointense bilateral posterior occipital SDH. At 26 days postnatal age (bottom row), left frontal subdural collections that do not conform to CSF signal intensity are present, consistent with spontaneous SDH. The patient had no history of trauma and had a negative evaluation for NAI."

More rarely, traumatic births may also cause symptomatic intracranial hemorrhage.

Intracranial hemorrhage

Retinal hemorrhage

Benign external hydrocephalus

Benign external hydrocephalus (BEH), also known as enlarged extraaxial spaces, enlargement of subarachnoid spaces, or many other terms, is a medical condition "associating a rapid increase in head circumference, combined with enlarged subarachnoid spaces as seen on neuroimaging—especially overlying the frontal lobes—and normal or only moderately enlarged ventricles" (Zahl 2011). It is often considered as a risk factor for the development of quasi-spontaneous subdural hemorrhage and retinal hemorrhage, although not by part of the SBS/AHT adherents. Some authors believe that a number of children with BEH are misdiagnosed with SBS/AHT, especially due to imaging similarities.

This figure shows MRIs of two different infants, one diagnosed with SBS/AHT (left), one with BEH and no concern for abuse (right).
Left: Fig. 2b from Flom 2016. "Abnormal screening head MRI sequences in 2.5-month-old boy with abusive head trauma and Glasgow coma scale score of 15. b Axial gradient recalled echo sequence demonstrates interhemispheric and left convexity subdural hemorrhage (arrow). This sequence was assessed as abnormal.".
Right: Fig. 2 from Vinchon 2010 showing a case of "spontaneous subdural hemorrhage" with no concern for abuse. "Case 7: 5-month male with preexisting idiopathic macrocrania (+3 SD), without other past medical history, who presented with a 10-day duration syndrome of raised intracranial pressure with increased head circumference (symptomatic macrocrania with discontinuity of the growth curve). Computed tomography showed a purely hypodense collection, which magnetic resonance imaging (axial view, T2-weighted) showed to be both subdural and subarachnoid. Other explorations were all negative. Subdural subtraction of 20 ml of bloody cerebrospinal fluid was followed with subduroperitoneal drainage and uneventful recovery. The coexistence of subdural hematoma (SDH) with arachnoidomegaly shows that both compartments are under pressure and suggests that preexisting hydrodynamic fragility was decompensated with the occurrence of an SDH."

SBS/AHT adherents


Differential diagnoses

Animal models

Obviously, shaking cannot be reproduced in infants in an experimental setting. Researchers have instead investigated shaking in animal models. To date, results have been inconclusive: the characteristic features of the triad, including bridging vein rupture, subdural hematoma, and retinal hemorrhage, have not yet been observed in animal models of shaking. On the other hand, traumatic cervical and brain injuries have been described.

"The experimental animal models, in both laboratory rodents and domestic animal species, developed to date were unable to reliably replicate the full spectrum of neuropathologic changes found in human infant AHT." (Finnie 2022)

Biomechanics and numerical modeling

Biomechanics is another research avenue, where physical or virtual models of a baby's head are developed and subject to various traumatic conditions. Kinematics and materials deformations can be measured and compared to known or estimated tissue injury thresholds.

Understanding of biomechanical properties of pediatric tissues remains elusive, whereas scaling of injury thresholds from animal models or adults to human infants is challenging.

Results have been inconclusive so far, and it remains unclear whether shaking alone can generate sufficient forces to cause traumatic rupture of bridging veins, subdural hemorrhage, and retinal hemorrhage. Forensic implications of such negative findings are also unclear due to intrinsic limitations of biomechanical and finite element models.

On the other hand, forces generated during impacts have been found to be consistently larger than with shaking alone.

"Although only Cory and Jones reported rotational head accelerations in excess of predicted injury thresholds for concussion, no study has to date demonstrated that shaking alone, without an associated impact, exceeds the injury thresholds associated with SDH." (Jones 2015, p. 292)

Figure 3 from Wilhelm 2020. "Figure 3. Development of a numerical model of a small child’s head depicted in five stages."

SBS/AHT adherents


Short falls

Severe injuries following short falls are rare, but how rare they are is controversial. To some authors, they are so exceptional that histories of short falls can be considered to be falsified in almost all severe or fatal cases. To other authors, these statistical arguments are not correct, and short falls should be considered as an important differential diagnosis of subdural and retinal hemorrhage in infants.

SBS/AHT adherents



Many authors from both sides agree that, since experiments cannot be conducted on babies, and since videotaped and witnessed shaking events are vanishingly rare, confessions are the main evidentiary basis of SBS/AHT. Adherents believe in the validity of most confessions of shaking, while agnostics claim they are unreliable due to the existence of guilty pleas and false confessions during police interrogation.

For a SBS/AHT adherent:

"To those who argue that the contribution of shaking to the pathophysiology of AHT is a hypothesis lacking a sufficient evidentiary base, the consistent and repeated observation that confessed shaking results in stereotypical injuries that are so frequently encountered in AHT—and which are so extraordinarily rare following accidental/impact injuries—is the evidentiary basis for shaking." (Dias 2011, p. 310)

For agnostics:

"Moreover, (...) it is unsettling that physicians would rely so heavily on purported perpetrator confessions as somehow validating questionable SBS/AHT beliefs. Approximately twenty-five percent of the DNA exonerations in this country were in cases where the innocent defendant had allegedly confessed (...). Many purported SBS "confessions" are obtained pursuant to plea agreements; or as a requirement to retain or regain parental rights in dependency proceedings; or the confession was merely to shaking the child in an effort to resuscitate after the child had already collapsed; or the confession was induced by law enforcement through suggestion that confessing to shaking would allow the child to receive important medical treatment or was the "only" possible explanation for the child's condition." (Papetti 2019, pp. 342-343)

SBS/AHT adherents


Epidemiology and prevention

Epidemiological studies investigate the demographics of children diagnosed with SBS/AHT, the incidence of these medical determinations, and the impact of prevention programs on the global incidence of determinations of abuse.

SBS/AHT adherents


Medical ethics, medicolegal issues, cognitive biases

SBS/AHT adherents