August 2009 | Back to Table of ContentsFace to FaceChallenging an AssumptionA pathologist questions shaken baby syndrome.By Kate LedgerFor John Plunkett, M.D., the case in 1986 that would put a new spin on his career was not unlikeothers he’d seen before. A general and forensic pathologist who occasionally consulted forattorneys, Plunkett was asked to review the post-mortem findings following the death of an 18-month-old girl.According to the mother, the baby had climbed onto the arm of a couch to reach for a figurine ona shelf, then fell, hitting her head on the linoleum floor. But physicians at MinneapolisChildren’s Hospital and the Hennepin County Medical Examiner didn’t believe the story. Thebaby showed bleeding inside the dura within the skull, hemorrhaging of the blood vessels in theretinas, and altered function in the brain. Those who’d examined her both before and after shedied saw no evidence of impact on her head. “They were convinced this could only be inflictedtrauma,” Plunkett recalls. What they believed, based on the specific injuries, was that the motherhad shaken her baby to death.As the mother went to trial, Plunkett began to wonder about the symptoms: Where was theevidence that subdural bleeding, retinal hemorrhaging, and brain swelling—considered theclassic triad of signs pointing to a shaken baby—added up to murder? His questioning of thepost-mortem findings spurred him onto a new tack of research, and he began to investigate theconcept of “shaken baby syndrome.” What he’s found over the years has turned him into acontroversial figure in cases of alleged child abuse. Today, physicians in pediatrics andpathology have wide-ranging opinions of the widely published pathologist, now retired from hishospital work, as he continues to consult, write, and speak on infant head injury. “He’s bothrevered and reviled,” says pathologist Susan Roe, M.D., who works for Regina Medical Centerand serves as a forensic pathologist for the Minnesota Regional Medical Examiner’s office. ButPlunkett has remained emphatic about his belief: “You can’t shake a baby to death.”Parsing the EvidenceJohn Jerome Plunkett was born in St. Paul and lived first in Highland Park and then in themiddle- and working-class Midway neighborhood. His father was a lawyer who spent almost hisentire career as a Ramsey County District Judge. His uncles were also attorneys, and hisbrothers and cousins went into the field as well. “I grew up with a real love for the law,”
Plunkett acknowledges. But while completing a bachelor’s degree in history and chemistry atthe University of Minnesota, he found himself drawn to medicine instead.After earning a medical degree in 1972, he realized he was interested most of all in askingcertain types of questions about patient cases, which led him to pathology and “the mostscientific part of medicine.” After an internship and pathology residency at St. Paul-RamseyMedical Center (later to become Regions Hospital), he completed a forensic fellowship at theHennepin County Medical Examiner’s Office. His plan was to be “a general pathologist and amedical educator first and a forensic pathologist second.”True to his career goals, Plunkett worked as a pathologist and director of education at what isnow Regina Medical Center in Hastings and spent nearly a decade as Hennepin County’s deputymedical examiner. As a forensic pathologist, his findings were often used in court, placing himat the intersection of science and the law, something he found “very appealing.”One trait Plunkett came to be known for was his willingness to wonder aloud. Pathologist JaniceOphoven, M.D., who trained concurrently with Plunkett and worked with him at the medicalexaminer’s office, says he was always comfortable questioning. “He would always ask, ‘What’syour proof for this?’ and he wasn’t afraid to challenge people.” Plunkett did just that in the 1986case in which the mother was convicted of second-degree murder. When Plunkett looked atautopsy photographs of the 18-month-old girl, he discerned a bruise on the skull where othershad seen none. “I said, ‘Wait a second. Why couldn’t this impact injury have caused what we’reseeing in the brain?’” he recalls. The response he received from other physicians was that short-distance falls can’t cause serious injury or death in an infant.But Plunkett was skeptical. He went back to the literature, looking for evidence. Two journalarticles, one American and one British, published in the early 1970s had established thediagnostic paradigm that would come to be known as shaken baby syndrome. In the absence ofany other signs of trauma, the papers stated, subdural hematoma, retinal hemorrhaging, and brainswelling were signs of forceful shaking. Furthermore, the papers stated, no other kind of traumacould cause the three symptoms.Digging into case studies of accidental deaths, however, Plunkett found incidents that suggestedotherwise. In 2001, he published an article in the American Journal of Forensic Medicine andPathology documenting 18 confirmed cases in which infants and young children died from fallsof less than 10 feet. “So clearly it can happen,” he says. At the same time, he began lookingclosely at literature from the automotive industry, in particular, studies of child crash-testdummies, where researchers had calculated the effects of acceleration and deceleration on thehuman body. With calculations of cycles per second (human hands can only shake a baby atabout four cycles per second) and how far the head moves, Plunkett explains, “using just plainold trigonometry you can calculate the acceleration [of the head], the change in velocity overtime.” Comparing that figure to known injury thresholds based on experimental results fromanimal studies, cadaver studies, or reconstruction of real-life accidents, it’s possible to determinethe type of forces that would cause injury. Even though the forces involved in shaking couldcause an injury such as whiplash, Plunkett states, “they’re nowhere even near the thresholdrequired to cause brain injury.”
He wasn’t alone in wondering about shaken baby syndrome. Across the country, a handful ofresearchers had begun questioning it. One was pediatric neurosurgeon Ann-Christine Duhaime,M.D., now at Dartmouth-Hitchcock Medical Center, whose research involved modeling headtrauma. Her studies showed that shaking would cause less than one-tenth the amount of forceneeded to inflict injury.To Plunkett’s thinking, a variety of other scenarios could result in the classic triad of symptoms.He suggests meningitis, encephalitis, or chronic subdural hematomas, for instance, a yet-undiagnosed inborn error of metabolism or “some natural disease.” In addition to writing journalarticles questioning the classic triad and presenting his findings at conferences, Plunkett begantestifying in court cases, opposing prosecution that used it to get murder convictions. What hehas emphasized is that the medical community still lacks sufficient proof for its definition ofshaken baby syndrome. In the absence of other injuries—when there are no head bruises orbumps, and no signs of old broken bones or bruises on other parts of a baby’s body—the classictriad does not mean shaking necessarily has taken place. He believes that the seminal papersfrom the 1970s misinterpreted the original research results of Ayub Ommaya, M.D., aneurosurgeon and biomechanical engineer, who had at one time been head of the neurosciencesbranch of the National Institutes of Health. Over the years, physicians’ refusal to re-examine thetenets about shaking in babies with no other signs of injury, has undoubtedly led to theconviction of innocent citizens. “People just believed this stuff for so long,” Plunkett says, “ittook on a life of its own.”A New Look at the IssueOphoven, who is now assistant medical examiner for St. Louis County and a specialist in childabuse and injury, remembers being put off at first when Plunkett began questioning shaken babysyndrome. Then, in the late 1990s, as more papers appeared questioning the symptoms, shebegan to review the literature and changed her mind. “The fundamental evidence has never beenscientifically validated,” she says. Since testifying for the defense in cases where no signs ofabuse are present but the baby shows the classic triad, even physicians have called her names,from “defense whore” to “wicked.” Ophoven adds, “John has been called worse.”Plunkett’s influence may not have pushed the majority of forensic pathologists to becomedefense advocates. But some may now be more conservative in their judgment, speculates Roe.“John has been one of the key people getting our whole community to relook at this wholeissue,” she says. Many forensic pathologists might be more inclined today to call a case with noother evidence “undetermined” rather than label it homicide, she adds.But pediatricians, many of whom say they like Plunkett personally, having met him in court or atconferences, have taken issue with his findings. Some have questioned the math in thebiomechanical engineering studies. Others have looked closely at studies such as Duhaime’s andcritiqued the modeling of a baby’s brain, pointing out that no material exists to replicate anewborn’s skull. Although many physicians concur that it’s positive to be asking questionsabout the validity of the diagnosis, ultimately “there is no credible medical evidence to supportthe notion that shaking does not cause these injuries,” says Robert Block, M.D., former chair ofthe American Academy of Pediatrics’ Committee on Child Abuse and Neglect. Block co-
authored a policy statement that was issued earlier this year by the AAP, formally changing thename shaken baby syndrome to abusive head trauma. (For public health messages to families,the term shaken baby syndrome will still be used.) The goal of the paper, says Block, was toshift the focus to the injuries and the incidence of abuse and away from the debate over whethershaking is the key mechanism.Without a doubt, the debate has prompted a closer look at the classic triad. In the last decade,notes pediatrician Rich Kaplan, M.D., of Children’s Hospital and Clinics of Minnesota and theUniversity of Minnesota, research has offered refined descriptions of some of the symptomsassociated with shaking. “For instance, there’s greater understanding of retinal hemorrhaging,and today, we know that a small amount of bleeding in the back of the retina can be caused bysomething other than abuse.” Kaplan adds that pediatricians have become better informed in thelast several years and now consider a wide range of possible causes; but he still worries that theopposite is of greater concern: Pediatricians and other health care providers are still not seeingabuse when it occurs.Even so, Plunkett maintains that the triad continues to be used so vigorously by prosecutors thatan accused person faces an uphill battle. “You basically have to prove your innocence,” he says,estimating that the number of people inappropriately convicted might be in the thousands. Evenwhen caregivers are on record for having “confessed,” he points out, their acknowledgment of“shaking” has referred to the moments of reviving a baby who was already unconscious or evenjostling a baby on a knee.What’s given him hope lately is that in the United Kingdom and Canada, old cases of abuse haverecently been reopened and reinvestigated with new awareness of the controversy about shakenbaby symptoms. In Britain, for instance, if the only findings are subdural hematoma, retinalbleeding, and brain swelling—with no other signs of trauma, no history of harm, and nowitnesses—they are no longer enough to bring charges. Plunkett believes that the United Statesis still far off in setting justice straight, but he hopes the medical world will eventually take noteof what’s “really a paradigm change” in approaching infant injury evaluation. “At the end of theday,” he says, “the default diagnosis is not abuse, it is ‘I don’t know.’” MMKate Ledger is a freelance writer in St. Paul, Minnesota, and a frequent contributor to Minnesota Medicine.
January 2010 | Back to Table of ContentsLettersEvidence Outweighs BeliefCongratulations on the publication of a significant and meaningful cover story featuring Dr.David McCollum and focusing on the hidden costs of abuse and neglect (August, p. 26). Articlesby Dr. Therese Zink (p. 32), Dr. Amy Walsh (p. 37), and Susan McCormick Hadley (p. 41) addto your appropriate emphasis on the health costs of violence and abuse.Unfortunately, your feature on Dr. John Plunkett falls short of the high caliber of the otherreports. One of our International Advisory Board members, Dr. Robert Block, spent a significantamount of time talking with reporter Kate Ledger, who chose to condense their longconversation into a single quote and a statement. The result is that the article is misleading andmisrepresents the spectrum of fact and medical opinion on the issue of shaken babies andabusive head trauma.Dr. Plunkett is well-known to the medical experts who frequently find their medical opinionschallenged by him as he tours the United States as a hired defense witness. He remains“emphatic,” says Ms. Ledger, “about his belief: You can’t shake a baby to death.”While he is obviously entitled to his belief as a physician and defense witness, he is alsoobligated to study, interpret, and acknowledge the myriad solid medical evidence that does notsupport his belief. The article cites Dr. Plunkett’s paper from 2001, correctly stating that hedocumented 18 confirmed cases of death from falls of less than 10 feet. But the article does notinclude the fact that these few cases were culled from more than 75,000 reports of playgroundinjuries severe enough to be reported, nor does it include the fact that there were no infantdeaths, and only five of the children were between the ages of 12 and 24 months. Most of thedeaths were not what most people would call “short falls,” as they involved playgroundequipment such as moving swings. His article actually confirms that deaths from short fallsamong infants and young children are extremely rare. A more recent peer-reviewed study findsthe incidence of death among young children experiencing short falls is in the range of 0.5 per 1million.Ms. Ledger reports that “Plunkett believes that the United States is still far off in setting justicestraight.” What a shame it would be for the memory of dead babies and for the parents of babieskilled by other caregivers if “setting justice straight” meant accepting a minority opinion totallywithout biomechanical, epidemiologic, or biological confirmation.
We find it concerning that the author failed to cite the enormous body of peer-reviewed medicalliterature from countries and centers around the world that consistently and repeatedly supportthe concept of shaken baby syndrome and has led to the formation of statements to that effect bymultiple major medical organizations, not the least of which is the American Academy ofPediatrics.1-14Randell Alexander, M.D., Ph.D.Ronald G. Barr, MDCM, FRCPCRobert Block, M.D., FAAPMary Case, M.D.David L. Chadwick, M.D.Brian Holmgren, J.D.Carole Jenny, M.D., MBAJohn M. Leventhal, M.D.Alex Levin, M.D., MHScRobert Reece, M.D.Philip Wheeler, DCIMarilyn Barr, executive directorInternational Advisory BoardNational Center on Shaken Baby SyndromeAbbreviated Reference List1. American Academy of Pediatrics Committee on Child Abuse and Neglect. Shaken baby syndrome: rotational cranial injuries-technical report.Pediatrics. 2001;108(1):206-10.2. Barlow KM, Milne S, Aitken K, Minns RA. A retrospective epidemiological analysis of non-accidental head injury in children in Scotland overa 15 year period. Scot Med J. 1998;43:112-4.3. Bechtel K, Stoessel K, Leventhal J, et al. Characteristics that distinguish accidental from abusive injury in hospitalized young children withhead trauma. Pediatrics. 2004;114:165-8.4. Block RW. Child abuse—controversies and imposters. Curr Probl Pediatr. 1999;29:253-72.5. Case ME, Graham MA, Handy TO, Jentzen JM, Monteleone JA. National Association of Medical Examiners Ad Hoc Committee on ShakenBaby Syndrome. Position paper on fatal abusive head injuries in infants and young children. Am J Forensic Med Pathol. 2001;22(2):112-22.6. Chadwick DL, Bertocci G, Castillo E, et al. Annual risk of death resulting from short falls among young children: less than 1 in 1 million.Pediatrics. 2008;121(6):1213-24.7. Christian CW, Block R. Committee on Child Abuse and Neglect. American Academy of Pediatrics. Pediatrics. 2009 May;123(5):1409-11.8. Jaspan T. Current controversies in the interpretation of non-accidental head injury. Pediatr Radiol. 2008;38(S3):S378-87.9. Keenan HT, Runyan DK, Marshall SW, et al. A population-based study of inflicted traumatic brain injury in young children. JAMA.2003;290:621-6.10. Kivlin JD, Simons KB, Lazoritz S, Ruttum MS. Shaken baby syndrome. Ophthalmology. 2000;107:1246-54.11. Minns RA, Brown JK. Shaking and other non-accidental head injuries in children. London: MacKeith Press, 2005.12. Morad Y, Kim YM, Armstrong DC, Huyer D, Mian M, Levin AV. Correlation between retinal abnormalities and intracranial abnormalities inthe shaken baby syndrome. Amer J Ophthalmol. 2002;134:354-9.13. Punt J, Bonshek RE, Jaspan T, et al. The “unified hypothesis” of Geddes et al. is not supported by the data. Pediatr Rehabil. 2004;7:173-84.14. Starling SP, Patel S, Burke BL, et al. Analysis of perpetrator admissions to inflicted traumatic brain injury in children. Arch Pediatr AdolescMed. 2004;158:454-8.1300 Godward St. NE, Suite 2500, Minneapolis, MN 55413 - Phone: (612) 378-1875 - Fax:(612) 378-3875We welcome your comments and suggestions about this site. Please contact email@example.com.The editors are pleased to consider for publication original research, reviews, case reports,essays, poetry, letters, and opinion pieces related to medical practice in Minnesota that have notbeen published or submitted elsewhere.
February 2010Letters to the EditorFlawed LogicI am writing in response to the letter published in your January 2010 issue (“Evidence OutweighsBelief,” p. 5), which was critical of Dr. John Plunkett’s views on the various inconsistenciesrelated to so-called shaken baby syndrome (SBS) (see article on Dr. Plunkett in the August 2009issue, p. 22).* There are many physicians who, after serious consideration and research, are quitecertain that SBS is invalid when applied to healthy children.Short falls do cause serious injury, and citing the infrequency of serious injury from such falls isa deception. Children who land on their head after falling from 3 feet are moving at 13.5feet/second and hit the ground with a force of 259 pounds. When the head hits a hard floor, 130gof deceleration (130 times the force of gravity where g is acceleration due to gravity) are appliedto the head. This is above the known threshold for injury, which is about 50g. Add 18 inches tothe distance of a fall (a baby standing on a counter), and the head would be traveling at 17feet/second and generating a force of 180g on impact.1 It is clear that there is substantial forceinvolved in short falls and that as the fall distance increases, the risk of injury increasesaccordingly.It is worth noting that a child who lands on his head after falling from those heights will beinjured and will need to be seen in an emergency department. Although such falls are infrequent,infrequency is not a basis for discounting the possibility of accidental trauma and, thus, not thebasis for discounting the accounts of apparently reliable caregivers.The deception perpetrated by the signers of this letter equates infrequency of a serious fall withthe improbability of an accidental fall. This is both a logic flaw and an intentional disregard ofthe physics of falling. Although the fact that one in a million falls will result in a serious headinjury may be true, it means that if a million children have accidental falls in a day, then therewill be a serious head injury from at least one of those falls every day. We know that number ishigher based on experience in emergency medicine. On the premise that only a small fraction ofaccidental falls are serious, the signers of this critique would discount the history of the caregiverand conclude that he or she has abused the child. In the absence of impact trauma to the head,intentional injury would be the presumption and SBS the default diagnosis.Reliable studies using sophisticated biofidelic instrumented crash dummies have shown shakingcan generate at most about 10g of force.2,3 Bandak has shown beyond dispute that neck damagewill result before brain injury can occur with shaking.4
In every case in which shaking is alleged without neck injury, which is virtually every case,shaking can be ruled out as the cause of primary intracranial trauma.Clinically, the misconception that common short falls do not cause serious injury was studied byGreenes and Schutzman. In an article published in the Annals of Emergency Medicine in 1997,they report that among children who fell less than 3 feet and were brought to an emergencydepartment by a concerned parent, approximately 18 percent had severe intracranial trauma orskull fracture.5 In another study published in 1998, the authors found 18 percent of children withsignificant head injury, primarily fractures, were without neurologic symptoms or evidence ofexternal trauma.6The use of retinal hemorrhage to diagnose SBS is flawed as well. In 1984, Aoki found retinalhemorrhage in 100 percent of children with subdural hematomas from impact trauma with noshaking.7The frequency of perinatal subdural hematoma evolving into chronic subdural hematoma(CSDH), with its inherent complications, while infrequent, is similarly discounted.8 A 0.5percent complication rate in the United States alone will yield 2,000 cases of CSDH presentingas neuropathic events in infants, which invariably are diagnosed as abuse.Furthermore, the scope of the research that led to the creation of SBS has been shown to beflawed by Donohoe. In an article published in the American Journal of Forensic Pathology in2003, he wrote that there was inadequate scientific evidence to come to “a firm conclusion onmost matters pertaining to SBS,” and he called for properly controlled, prospective trials intoSBS. “Without published and replicated studies of that type,” he wrote, “the commonly heldopinion that the finding of subdural hematoma and retinal hemorrhage in an infant was strongevidence of SBS was unsustainable, at least from the medical literature.” 9The investment in maintaining the mythology of SBS is substantial for the signers of this letterand others. Many people are in jail, and many families have been ripped apart based on theirefforts. The injustice caused by the application of nonevidence-based theories to prosecuteinnocent caregivers is real, and many courts in the United States and the government of Canadahave acknowledged this.10I agree with the title given to this critique. The evidence does outweigh belief. But in this case,the evidence trumps the unsupported belief in this improbable and unproven construct ofshaking.Lastly, every physician and law- enforcement professional should be aware that there has neverbeen a witnessed or videotaped incident of shaking a healthy child in which the child wasimmediately examined and found to have either retinal hemorrhage or subdural hematoma.Author:Steven Gabaeff, M.D., F.A.A.E.M.Emergency and Clinical Forensic MedicineSacramento, California
Cosigners:Thomas L. Bohan, Ph.D. (physics), J.D.President, American Academy of Forensic Sciences (as an individual)Marvin Miller, M.D.Professor of Pediatrics and Ob/GynAffiliated Professor of Biomedical Engineering,Wright State University Boonshoft School of MedicineDayton, OhioHorace Gardner, M.D.OphthalmologyManitou Springs, ColoradoRobert K. Rothfeder, M.D.Emergency PhysicianSalt Lake City, UtahJoseph Scheller, M.D.Forensic PathologyAssistant Professor of Paediatrics and Neurology,George Washington University School of MedicineWashington, D.C.David M. Posey, M.D.Glenoaks Pathology Medical Group, Inc.American Autopsy Services, Inc.References: 1. The chain of physics calculations used is this: Impact velocity, V, resulting from free fall from a given height, h, is: V = SQRT(2*g*h). Average acceleration at impact, a, occurring over a time, dt, is a = (dV/dt)OR, average acceleration, a, occurring over a stopping distance, d, is: a = V^2/(2*d). For a 3-foot fall (h = 3 feet), V = 13.9 ft/sec. If we pick a reasonable impact duration of 7.5 msec, then: a = (13.9 / 0.0075) = 1,853 ft/sec^2 or 58g and for a velocity of 17 feet/sec then a = (17 / 0.0075) = 2,666 ft/sec^2 or 70g. Another approach is to pick a reasonable stopping distance of 0.5 inches (0.042 ft) then: a = 13.9^2 / (2*0.042) = 2300 ft/sec^2 or 71g at 3 feet and 106g at 4.5 feet. If we split the difference between the two approaches, that is roughly 65g average acceleration at 3 feet and 90g at 4.5 feet. If we assume a triangular force pulse at impact (standard concept), then the peak values are double the average or about 130g peak linear acceleration from 3 feet and 180g at 4.5 feet. The IARV (injury assessment reference value) for the CRABI 6 month ATD is 50g. (Calculations validated by Kirk Thibault.) 2. Commonwealth v Ann Power, 2005. Report to the Middlesex County District Attorney’s Office, Cambridge Massachusetts by Carole Jenny, dated December 29, 2005. Letters 319.
3. Prange MT, Coats B, Duhaime AC, Margulies SS. Anthropomorphic simulations of falls, shakes, and inflicted impacts in infants. J Neurosurg. 2003;99(1):143-50.4. Bandak FA. Shaken baby syndrome: A biomechanics analysis of injury mechanisms. Forensic Sci Int. 2005;151(1):71–9.5. Greenes DS, Schutzman SA. Infants with isolated skull fracture: what are their clinical characteristics, and do they require hospitalization? Ann Emerg Med. 1997;30(3):253-9.6. Greenes DS, Schutzman SA. Occult intracranial injury in infants. Ann Emerg Med. 1998;32(6):680- 6.7. Aoki N, Masuzawa H. Infantile acute subdural hematoma: clinical analysis of 26 cases. J Neurosurg. 1984;61(2):273-80.8. Rooks VJ, Eaton JP, Ruess L, Petermann GW, Keck-Wherley J, Pedersen RC. Prevalence and evolution of intracranial hemorrhage in asymptomatic term infants. AJNR Am J Neuroradiol. 2008;29(6):1082-99. Donohoe M. Evidence-based medicine and shaken baby syndrome, part I: literature review, 1966– 1998. Am J Forensic Med Pathol. 2003;24(3):239–42.10. Cordner S, Ehsani J, Bugeja L, Ibrahim J. Pediatric forensic pathology: limits and controversies. Commissioned by the Inquiry into Pediatric Forensic Pathology, Ontario, Canada; November 28, 2007. Available at: www.goudgeinquiry.ca/policy_research/pdf/Limits_and_Controversies- CORDNER.pdf. Accessed January 11, 2010.