Shaken Baby Syndrome

Penal Code §273ab Assault on Child With Force Likely to Produce Great Bodily Injury Resulting in Death–Punishment.


1. The Original Hypothesis

     In 1962, Dr. Henry Kempe wrote a very influential article identifying characteristics of “battered” children. He listed several physical injuries that, particularly when more than one was present, were suspicious for child abuse. Most were fairly common sense–broken bones in babies, soft tissue swelling, bruises. Also on his list was subdural hematoma– a pooling of blood between the brain itself and the protective dura layer, which Dr. Kempe identified as often a traumainduced injury.

     In 1971, Dr. A. Norman Guthkelch, who was the first pediatric neurosurgeon in England, wrote an article that questioned why infants who presented with subdural hematoma and who he suspected had been abused, nevertheless did not have any sign of trauma to their heads. He cited a biomechanical study by a Dr. Ommaya that Dr. Guthkelch described as recording “two welldocumented cases of subdural haematoma, in both of which the subject sustained a whiplash injury to the neck as a result of an automobile accident, the head itself not being [impacted] at all.” He also discussed two patients of his that had subdural hematomas yet no sign of head trauma – in one the mother said she had shaken her infant when he was having a coughing fit and she feared he was choking; in the other, the infant had grip marks and the mother said that she “might have” shaken him when he cried at night. From the Ommaya study and his two case reports, Dr. Guthkelch hypothesized that infants could sustain whiplash-type injuries, including subdural hematoma, from being violently shaken.

     In 1972 and 1974, a prominent American pediatric radiologist and textbook author named John Caffey published two articles, respectively titled: On the Theory and Practice of Shaking Infants: Its Potential Residual Effects of Permanent Brain Damage and Mental Retardation, and The Whiplash Shaken Infant Syndrome: Manual Shaking by the Extremities With WhiplashInduced Intracranial and Intraocular Bleedings, Linked with Permanent Brain Damage and Mental Retardation. In the first article, Dr. Caffey collected 27 instances of what he deemed “convincing” examples of children who had suffered brain injury as a result of shaking. In the second article, Dr. Caffey cited his previous data and the same Ommaya study that Dr. Guthkelch cited for the proposition that shaking infants could cause subdural hemorrhage. Dr. Caffey speculated that, in addition to causing subdural hemorrhage, shaking damaged capillaries within the retina, which explained why retinal hemorrhages often were seen in children he thought to have been shaken. Although Dr. Caffey admitted that his data set was “meager” and “manifestly incomplete,” he broadly concluded that the evidence “indicates that manual whiplash shaking of infants is a common primary type of trauma in the so-called battered infant syndrome. It appears to be the major cause in these infants who suffer from subdural hematomas and intraocular bleedings.”

2. SBS Rapidly Becomes A Well-Accepted Medical Diagnosis

     Notwithstanding that Dr. Caffey reached his conclusions on an evidence base that even he acknowledged was meager, but propelled by a nationwide campaign highlighting the dangers of shaking infants, the SBS diagnosis rapidly gained acceptance in medical circles. See Uscinski, Shaken Baby Syndrome: An Odyssey (“Nonetheless, the mechanism of shaking and the so named syndrome gained immediate acceptance and enormously widespread popularity, with no real investigation or even question as to its scientific validity.”) Immwinkelried, Shaken Baby Syndrome: A Genuine Battle of the Scientific (and Non-Scientific) Experts, (“in a relatively short time after Caffey’s enunciation of the theory, the theory became widely accepted in both medical and legal circles.”). Shaken Baby Syndrome was not always defined consistently in the literature – for example, it often was applied in cases where there was evidence of impact to the head as well as in cases where there was not. But the general theory is typically expressed as shaking causes the brain to move within the skull, which, in turn, causes bridging veins overlying the brain to rupture and tear, which, in turn, causes blood to form within the subdural area between the brain and the overlying protective dura. Consistent with Dr. Caffey’s hypothesis, the same accelerationdeceleration mechanism is assumed to cause capillaries within the retina to shear and hemorrhage.

     By the early 1990s, SBS – a diagnosis that an infant who presented with subdural hemorrhage, retinal hemorrhages and no “adequate” explanation for such allegedly traumatic injuries presumptively had been violently shaken or slammed – was an entrenched diagnosis within the medical community.

     Because SBS, by its very definition, is a diagnosis of violent shaking, it basically also is a diagnosis of child abuse. Consequently, if the baby died, an SBS diagnosis is, in essence, “a medical diagnosis of murder.”

     As the SBS diagnosis became more and more entrenched, SBS-based prosecutions and child protective services proceedings became common. By 2000, a National Center for Shaking Baby Syndrome led by a board of prominent physicians had been established to host conferences, distribute educational literature, train law enforcement officers, and support prosecutors in SBS cases. Manuals were published to guide prosecutors in SBS cases. See, e.g., Holmgren, Prosecuting the Shaken Infant Case in THE SHAKEN BABY SYNDROME: A MULTI DISCIPLINARY APPROACH 307 (2001) (providing prosecutors with ideas for physician testimony such as: the “expert can testify that the forces the child experiences [from shaking] are the equivalent of a 50-60 m.p.h. unrestrained motor vehicle accident, or a fall from 3-4 stories on a hard surface”). Thousands have been convicted.

3. SBS Reaches Peak Acceptance, Then Slowly Starts to Unravel

     Over the last decade, opposition to SBS has grown from a trickle to a virtual avalanche. The summary below provides a snapshot of this development:


     “The shaking hypothesis . . . was seemingly accepted as settled science in 2001 in two documents: a position paper from the National Association of Medical Examiners and an updated position statement from the American Academy of Pediatrics (AAP).” Lloyd, et al., Biomechanical Evaluation of Head Kinematics During Infant Shaking Versus Pediatric Activities of Daily Living. The AAP position statement endorsed SBS and suggested that child abuse be presumed whenever a child presented younger than one year with intracranial injury and retinal hemorrhages. The paper from the National Association of Medical Examiners (NAME) also endorsed SBS as a reliable diagnosis.

     Without dissent, numerous court decisions around the country at this point in time recognized SBS as a valid scientific theory upon which convictions could be sustained. In the 1980s and 1990s, dozens of articles presuming the existence, validity and reliability of the SBS diagnosis filled the medical literature.

     While SBS acceptance was at its zenith in 2001, hindsight reveals that the foundation for subsequent challenges to the SBS dogma were published that same year. In 2001, Dr. Jennian Geddes, a British neuropathologist, published two papers after studying the brains and eyes of infants who allegedly were the victims of non-accidental head injury, including shaking. In one of the papers, she observed that the subdural hemorrhage and brain findings in infants who died of natural causes appeared to be virtually indistinguishable from the findings in cases of allegedly abused children. In the other, she noted that, although it was assumed in SBS cases that a particular kind of shearing brain injury occurred that was, by definition, traumatic, she found no evidence of such shearing injury in studying the brains of babies thought to have been SBS victims. She concluded that the beliefs that shaking directly caused the triad (subdural hemorrhage, retinal hemorrhage, and edema) “require fresh examination.” At the time, her work was vilified as unreliable by the child abuse protection community.

     Also in 2001, John Plunkett, a forensic pathologist in Minnesota, published Fatal Pediatric Head Injuries Caused By Short-Distance Falls. In that article, Dr. Plunkett addressed common courtroom testimony that the triad could not be caused by falls unless the falls were from greater than 10 feet. Based on case data from the Consumer Product Safety Commission, he described multiple witnessed short falls that resulted in some or all of the triad injuries, including a videotaped fall of a 23 month-old toddler from a plastic gym set (28 inches high) on the carpetcovered garage of her home. The child cried and talked after the fall, but soon vomited, became stuporous and eventually died. The hospital findings included the SBS symptoms of subdural hemorrhage, retinal hemorrhage, and cerebral edema – all from a 28-inch fall.


     In 2002, Dr. Ayub Ommaya and heavyweight co-authors in the field of biomechanics published a lengthy article titled Biomechanics and Neuropathology of Adult and Paediatric Head Injury. Biomechanical engineers, unlike most medical doctors, study the exertion of forces on the human body and the body’s tolerances to such forces.

     In their article, Dr. Ommaya and his co-authors explained that Dr. Ommaya’s earlier whiplash study, the one that Guthkelch and Caffey cited in their seminal papers on SBS, had involved not infants, but adult rhesus monkeys. The monkeys had not been shaken, but instead had been strapped in collision carts and impacted at various speeds from the rear in an effort to gauge human thresholds to whiplash injury in car accidents. They further explained that the Ommaya study actually showed that subdural hemorrhage was caused far more easily by impact to the head than by whiplash and they suggested that the study had been misinterpreted by Guthkelch and Caffey in citing to it as scientific support for SBS.

     With respect to their views on SBS itself, they reasoned that they would expect to see soft tissue injury to the neck as well as spinal injury in any case of shaking sufficient to cause subdural and retinal hemorrhage. On the subject of retinal hemorrhages, they were directly critical of SBS theory, stating that the “hypothesis” of “retinal hemorrhage caused by orbital shaking has not been tested experimentally” and the “levels of force required for retinal bleeding by shaking to damage the eye directly is biomechanically improbable.”


     In about 1999, the medical community embraced a movement to ensure that medical practice was based on the best available medical and scientific evidence, as opposed to over reliance on anecdote and historical practice. This movement was known as the Evidence-Based Medicine (EBM) movement, and it developed repeatable criteria to gauge the evidentiary basis for medical practices and opinions, with Level I being the highest/most reliable evidence and Level IV the lowest/least reliable.

     In a 2003 article, Dr. Mark Donohoe classified the medical and scientific SBS literature through 1998 against EBM standards. His conclusions were startling. Although there were 55 published articles on SBS, none exceeded Level III-2 by the end of 1998, “which means that there was inadequate scientific evidence to come to a firm conclusion on most aspects of causation, diagnosis, treatment, or any other matter pertaining to SBS.” Dr. Donohoe concluded that “there was an urgent need for properly controlled, prospective trials into SBS, using a variety of controls. Without published and replicated studies of that type, the commonly held opinion that the finding of SDH and RH in an infant was strong evidence of SBS was unsustainable, at least from the medical literature.”


     Over time, as more literature confirmed cases of retinal hemorrhages in a wide variety of circumstances where no abuse had occurred, SBS advocates increasingly began to claim that, although retinal hemorrhages may be found in circumstances unrelated to abuse, certain types of ocular or retinal hemorrhage were virtually always diagnostic for abuse. In 2004, however, Dr. Patrick Lantz published a case report finding perimacular retinal folds, retinal and optic sheath hemorrhage – findings that previously had been considered diagnostic of SBS/abuse – in a child hurt when a television tipped over and hit him on the head. Although the article involved a single case, Dr. Lantz reviewed the existing literature that claimed such ocular findings were diagnostic of SBS and concluded that the literature suffered from the same systemic deficiencies noted by Dr. Donohoe with respect to SBS in general.

     In a letter to the Editor of the journal Pediatrics written the same year, Dr. Lantz stated that the “vested dogma” that the trauma of shaking causes retinal hemorrhages “is a faith-based assumption, not a scientific fact.”


     As noted, in 2002 Dr. Ommaya and his co-authors had suggested that it was improbable that one could shake an infant hard enough to cause intracranial injuries without also causing significant neck and spinal injuries. In 2005, Dr. Faris Bandak, a biomechanical engineer, published a study after investigating that exact hypothesis. Dr. Bandak’s study confirmed that the level of force required to shake a healthy infant hard enough to produce subdural injury would in fact exceed the tolerance of the infant neck, causing near or total neck failure.(“Head acceleration and velocity levels commonly reported for SBS generate forces that are far too great for the infant neck to withstand without injury.”) His article thus seriously called into question the assumption that shaking alone could cause the triad of injuries associated with SBS, at least without significant neck or spinal injury This was a critical study, because such neck and spinal findings are conspicuously absent in SBS cases.

     As these evidence-based contributions to the medical and scientific literature began to build, SBS advocates dismissed them as failing to acknowledge the literature establishing that a multitude of caretakers over the years had confessed to causing the child’s injuries through violent shaking. Perpetrator confessions, the SBS advocates contended, proved the validity of the diagnoses. But when Dr. Jan Leestma, a neuropathologist at the Children’s Memorial Hospital at Northwestern University, closely examined the so-called SBS confession literature, he found that in the vast majority of the “confession” cases there was clear evidence of impact injury to the head – i.e., the child’s injuries likely had not been caused by shaking at all or, at least, were likely partially attributable to an impact. He found that the confession literature only recorded 11 “pure” shaking cases and several of those were questionable because no details were given about the degree or duration of shaking, or about the circumstances surrounding the confession. For example in some of the cases where the caretaker admitted shaking the infant, it turns out the “admission” was bouncing the baby during play or attempts to revive the baby when it was found unconscious. Dr. Leestma concluded that “confessions” did not provide an adequate basis to establish the reliability of the SBS diagnosis.


     In 2006, the National Association of Medical Examiners officially withdrew its 2001 position paper on SBS. At its annual meeting, presentations were made with titles such as Where’s the Shaking?: Dragons, Elves, the Shaken Baby Syndrome and Other Mythical Entities and The Use of the Triad of Scant Subdural Hemorrhage, Brian Swelling, and Retinal Hemorrhages to Diagnose Non-Accidental Injury Is Not Scientifically Valid.

     In a follow-up article the year before on confessions, Dr. Leestma lamented that the medical community’s acceptance of SBS theory had resulted in a lack of studies into other potential causes of the SBS triad of findings:

“It should be apparent that fromvirtually every perspective many flaws exist in the theory that shaking is causative. No case studies have ever been undertaken to probe even a partial list of possible confounding variables/phenomena, such as the presence of intracranial cysts or fluid collections, hydrocephalus, congenital and inherited diseases, infection, coagulation disorders and venousthrombosis. . . or recent or remote head trauma. Until and unless these and probably many more factors are evaluated, it is inappropriate to select one mechanismonly and ignore the rest of the potential causes.”


     Echoing Dr. Leestma’s call for greater consideration and investigation into other conditions that would mimic the SBS findings, Dr. Patrick Barnes compiled and published a lengthy paper that included a five-page summary of known non-traumatic causes that mimicked SBS.


     Despite these advances in the medical and scientific literature that served to undermine SBS theory, SBS prosecutions continued seemingly unabated, with at most passing recognition that the SBS theory had become at all controversial. An abrupt change came in 2008.

     In Ontario, Canada, there had been several documented, publicized instances of mistakes, wrongful accusations and even wrongful convictions in childhood death cases in Ontario, with a particular focus on cases involving the Hospital for Sick Children, in Toronto. This led the Ontario government to establish The Inquiry Into Pediatric Forensic Pathology in Ontario. Ontario Court of appeals Justice Stephen Goudge was appointed as its Commissioner. Commissioner Goudge held hearings and gathered evidence for more than a year before issuing his lengthy report on October 1, 2008.

     The Report observed that “one of the deepest controversies surrounding pediatric forensic pathology concerns shaken baby syndrome.” The Report noted the “evolution in forensic pathology in this area” had progressed such that “the predominant view is no longer that the triad on its own is diagnostic of SBS. Instead, the issue is fraught with controversy.” The Report went on to conclude that “our systemic examination has identified this particular area of forensic pathology as one where change has raised the real possibility of past error.” Commissioner Goudge called for a review of SBS convictions from 1986-2006 because “[t]he significant evolution in pediatric forensic pathology relating to shaken baby syndrome” and “the concern that, in light of the change in knowledge, there may have been convictions that should now be seen as miscarriages of justice.”” Ontario undertook that review, which is ongoing.

     Also in 2008, the Wisconsin Court of appeals granted post-conviction relief to a woman who had been convicted in 1996 of murdering an infant in her care. The court did so because “a significant and legitimate debate in the medical community has developed in the past ten years over whether infants can be fatally injured through shaking alone . . .and whether other causes may mimic the symptoms traditionally viewed as indicating shaken baby or shaken impact syndrome.”


     The Committee on Child Abuse and Neglect of the American Academy of Pediatrics has long been dominated by staunch SBS advocates. In 2009, the Committee nevertheless felt compelled to update its 2001 policy statement, reasoning that “advances in the understanding of the mechanisms and clinical spectrum of injury associated with abusive head trauma compel us to modify our terminology to keep pace with our understanding of pathologic mechanisms.” The Committee continued to insist that the confession literature supported shaking as a mechanism of injury, but nevertheless recommended that physicians use the term “abusive head trauma” (AHT) rather than Shaken Baby Syndrome, a tacit admission that the SBS diagnosis and mechanism of shaking had become highly controversial.


     By 2010, a debate was raging about SBS, yet there was growing consensus that: (1) brain swelling previously thought attributable to neurons sheared from shaking was actually the result of hypoxia (lack of oxygen to the brain) from whatever cause, and (2) there are many non-traumatic causes of subdural hemorrhage. But SBS (now “AHT”) advocates insisted that retinal hemorrhages are a reliable marker of child abuse, particularly if they were multi-layered, extended out to the ora serrata and/or were accompanied by optic nerve sheath hemorrhage. But the retinal hemorrhage hypothesis was severely undermined in February 2010.

     Unlike most medical examiner’s offices, the Dallas Medical Examiner’s Office routinely removed eyes from corpses for evaluation by consulting ophthalmologic pathologists. In order to assess the hypothesis that certain eye findings were associated with child abuse and SBS, the office studied the eyes and records in cases involving deceased children. On February 24 , 2010, Dr. Evan Matshes reported on the study. He explained that “[f]or many years, the dogma of pediatric forensic pathology was ‘retinal and optic nerve sheath hemorrhages are pathognomonic of abuse head injury,’including shaken baby syndrome. Growing controversy surrounding the existence of SBS led to questioning of that dogma.” The study revealed that retinal hemorrhages are commonly found in natural and accidental deaths, as well as homicides, and identified a statistically significant relationship between retinal and optic nerve sheath hemorrhage and the restoring of a perfusing cardiac rhythm following advanced life support and cerebral edema, regardless of etiology. In other words, where there is hypoxia, increased intracranial pressure and prolonged resuscitation efforts, retinal hemorrhages of all kinds follow; such hemorrhages are not diagnostic of nor caused directly by shaking. The study concluded that eye evaluations are of “limited value” in child death investigations.

     In 2010, Rubin Miller, a biomechanical engineer, and Marvin Miller, a pediatrician and geneticist, published an article that noted that male babies were diagnosed as victims of SBS and traumatically inflicted brain injury much more frequently than females. The authors also noted that by a very similar margin male babies more frequently suffered subdural hemorrhage from non-SBS causes. The authors strongly criticized the evidentiary basis for SBS and explained why male babies can be expected to suffer intracranial bleeding from non-traumatic causes. They recommended that less focus be given to trying to support the failed SBS construct and more attention be given to better understand why boys are more susceptible to the variety of conditions that mimic SBS.


     As previously mentioned, Dr. Waney Squier is a neuropathologist and lecturer at Oxford. In a recent article she reviewed the status of the SBS science:

     • SBS Is Not Proven By Either Confessions or Witnessed Shakings The SBS literature contains only three published reports of witnessed shakings. All three infants were already collapsed before the shaking event. Despite clear evidence in the literature that confessions are not reliable basis for validating SBS, SBS advocates nonetheless continue to rely heavily on such “confessions” as “proof” of the shaking hypothesis.

     • Shaking Does Not Generate Enough Force to Cause Intracranial Injury Biomechanical tests done over the course of nearly two decades have confirmed that the forces generated by shaking are: (1) insufficient to cause whiplash intracranial injury and (2) less than those the head endures from an impact after a short fall. Accordingly, “shaking is no longer a credible mechanism” for the SBS findings.

     • There Are Many Non-SBS Causes, or “Mimics,” of SBS Symptoms The differential diagnosis of a baby presenting with the SBS triad is now “wide.” It includes alternative explanations that “are often overlooked,” particularly cortical vein and/or central venous thrombosis (CVT). CVT often presents with symptoms such as “lethargy, poor feeding, vomiting or seizures.” Similarly, physicians often fail to diagnose early non-traumatic subdural bleeding (from whatever cause) because the symptoms are “non-specific,” such as vomiting, irritability, progressive enlargement of the head and, “ultimately, a seizure.”


     Deborah Tuerkheimer, law professor at Northwestern University, published Flawed Convictions: “Shaken Baby Syndrome” and the Inertia of Injustice (Oxford University Press, 2014), based on her exhaustive and in-depth study of SBS cases, both published and unpublished, throughout the United States. Flawed Convictions exposes extraordinary failings in the criminal justice system’s treatment of what is, in essence, a medical diagnosis of child abuse or murder. Tuerkheimer’s book highlights the fundamental inadequacies in the legal response to SBS’s sciencedependent prosecutions.

     The defense of Shaken Baby Syndrome, or “AHT” cases, requires an attorney with substantial experience and education in the medical issues that form the basis of the prosecution. And these cases requires extensive use of qualified medical experts.



All submissions are confidentially reviewed by
Brian J. White, Certified Criminal Law Specialist.

Practice Areas

When your freedom is at stake, having an Experienced Criminal Law Specialist on your side can make a difference.