The Agonizing, Complicated, Lingering Questions of Shaken Baby Syndrome

[Alison J Stevens]

The Agonizing, Complicated, Lingering Questions of Shaken Baby Syndrome

Ten years after Matthew Eappen’s death – for which nanny Louise Woodward was convicted of involuntary manslaughter – a doctor who treated the infant has reversed his position on the case. Why is so much still unknown about shaken baby syndrome?

By MARY CARMICHAEL  |  February 11, 2007

On February 9, 1997, suffering from massive brain trauma after apparently being shaken and slammed against a hard surface, 8½-month-old Matthew Eappen died at Children’s Hospital Boston. It was the beginning of a story that would capture the nation’s attention and eventually see the dramatic release of au pair Louise Woodward, first convicted of second-degree murder, then let go on a lesser charge after 279 days in prison. Holed up in a Boston hotel after her sentence was commuted, a then 19-year-old Woodward sent out a statement saying she thought medical science would eventually clear her name. “I am confident that the future direction of this case, including the ongoing efforts by my lawyers to investigate further the science underlying the case, will further justify Judge [Hiller] Zobel’s decision,” the statement read. “I pray that further investigation into the scientific evidence convinces the Eappen family that I did their son no harm.”

Ten years later, Sunil and Deborah Eappen are not convinced. Neither are most physicians. Dr. Robert Reece, a Tufts University pediatrician who is one of the nation’s most respected authorities on shaken baby syndrome, says, “The information that was presented at trial on the prosecution side was correct” – an opinion that lines up with those of the vast majority of doctors who study SBS.

But Woodward was prescient in one respect: There has been a groundswell of scientific research since her trial, and some of it calls into question the facts of the syndrome. Deborah Eappen says the research can offer people an all-too-easy “defense mechanism,” a way to avoid contemplating the horror of child abuse by explaining it away as something else. It can be a literal defense mechanism, too; many of those who doubt the diagnosis regularly testify in court. The arguments, however, have persuaded a small but vocal minority of doctors, including a radiologist who testified against Woodward and some who have never testified at all. At times, their debates with mainstream pediatricians can devolve into outright fights, the sort seen often in court but less so in the usually placid prose of medical journals. The bickering can seem like little more than he-said-she-said. But it may be rooted in something more significant: an argument about what constitutes evidence, in both medicine and law; why those two standards sometimes differ; and, when they do, which should prevail.

The debating experts, sadly, have plenty of victims to study. Among the most recent headlines: In September, a Framingham man was found guilty of involuntary manslaughter in a shaken baby case, and in late December, a potential case emerged after a 9-week-old boy from Dracut died. The National Center on Shaken Baby Syndrome estimates that US hospitals handle 1,300 cases of abuse-related head trauma in children every year, most of them diagnosed as SBS and many caused by parents or caregivers frustrated over incessant crying or fussiness. Research by Dr. Carole Jenny, director of the Child Protection Program at Hasbro Children’s Hospital in Rhode Island, suggests that many more cases go unnoticed by the legal system – and even by doctors trained to recognize them.

Why can some perpetrators shake babies without being found out? Dr. Robert Block, who chairs the American Board of Pediatrics’ group on child abuse, explains that the syndrome is not always accompanied by obvious outward signs. “The symptoms can range from instant death to very minor things, and we’re not really clear on why there’s such a variety.” Some critically injured shaken babies have broken ribs from being squeezed, but others, even if shaken roughly and repeatedly, may enter emergency rooms with far fewer telltale symptoms – lethargy, decreased appetite, difficulty breathing, and bleeding and swelling under the scalp – that can mimic medical conditions unrelated to abuse. In less serious cases, which can initially manifest as vomiting and mild seizures, shaken baby syndrome is even easier to miss. “We have one [shaken baby] in the hospital right now who presented with a seizure thought at the time to be due to a fever and an apparent ear infection,” says Block, who sees cases as Oklahoma’s chief child abuse examiner. The infant improved a few days later. But when he had another seizure, the hospital referred him to specialists, who found injuries that suggested abuse.

Without bruises, fractures, or accounts from witnesses, doctors must often rely on two clues to diagnose shaken baby syndrome, and both appear only under close examination. The first, a distinct type of retinal bleeding, is a sign that Deborah Eappen, an ophthalmologist, saw for herself while her son lay in intensive care. A baby’s eye muscles are weak, so the eyeballs rotate and rock back and forth when the body is shaken. The thick fluid inside the eyeballs pulls on the delicate, multilayered retinal tissue, causing it to pucker and bleed. More than 90 diseases can cause retinal bleeding, but the hemorrhages in SBS are scattered over the retina all the way to the edges. Though doctors must still look for rare disorders that might be implicated, the hemorrhages appear in “a specific pattern,” says Block, “that we just don’t see in other conditions.”

Brain scans and autopsies provide doctors with the second clue in suspected shaken baby cases. They tell the story of what most doctors think happened under Matthew’s crown of black curls. In the brain, the syndrome is not a single injury, but a cataclysmic cascade of events that begins as the gelatinous gray matter bangs against the inside of the skull. The strain shears the fragile walls of tiny blood vessels crossing the brain’s surface, and the branches that connect the nerve cells are stretched to the breaking point. If cells in the brain stem are among those damaged, the body may quickly lose control over heart rate, respiration, blood pressure, and temperature. The baby’s breathing then falls off tempo, depriving the brain of oxygen – a stress the organ can bear only for several minutes before it starts to swell. Pressed against bone, the nerve cells now begin to implode, leaking chemicals into the brain tissue that are toxic outside the cell walls. Once this breakdown begins, the brain cannot resist, and it fails.

Only a quarter of SBS cases have such tragic results. But of the patients who survive, very few make full recoveries. Had Matthew Eappen lived, he might have spent the rest of his life in a permanent vegetative state, blind, or plagued by massive learning disabilities and emotional disturbances. As it was, he died five days after he was admitted to the hospital. His death was not the end of his story, though; in some ways, it was the beginning.

The Matthew Eappen/Louise Woodward trial changed society in ways that are still reverberating today. It pushed a little-known and poorly understood syndrome into the spotlight, ensuring that fewer people could plausibly claim ignorance of the dangers of violent shaking, as opposed to gentle bouncing – and that more of them would be prosecuted. Toni Blake, a top SBS defense lawyer based in San Diego, picked up her first case while the verdict in the Woodward trial was still pending. She says the prosecutions “exponentially increased” after that, growing from about 200 a year nationwide to more than 1,000 today. The trial, as Woodward had hoped, also set off a wave of new research. “That is the one good thing that came out of it,” says Reece, the Tufts pediatrician. “It made everybody aware that there were medical questions that needed further explanation.”

One of those who set off in pursuit of explanations was Dr. Patrick Barnes, a radiologist who was working at Children’s Hospital Boston at the time. Barnes treated Matthew Eappen and testified in the Woodward case for the prosecution. In the following years, though, he made a turnaround in his views that left many of his colleagues astonished. He now regularly testifies for defendants in SBS cases, arguing that caregivers can be wrongly blamed and sometimes sent to prison when rare medical conditions and previous injuries cause bleeding in infants’ brains. Speaking about Barnes, Reece says, “I don’t know what’s going on in his brain – it’s a mystery to all of us.”

Barnes says he had been indoctrinated throughout his career to believe that “there were certain features that were absolutely classic, that they were child abuse and could be nothing else.” He didn’t question those teachings, he says, until soon after the Woodward trial when he learned more details of the case, such as that Matthew had a weeks-old wrist fracture. “None of this was ever revealed to me by the prosecution,” he says. “I began to have real concerns about how a case like that was being handled.” He also began to wonder if previous injuries or genetic vulnerability could have caused the fatal insult to Matthew’s brain.

Barnes delved into the medical literature on head trauma, neurosurgery, and biomechanics. At Children’s Hospital, he began to examine infants and images brought to his attention by defense attorneys in other cases. “I started seeing a number of cases and recommending that we do MRIs,” he says. “We saw other conditions – accidental trauma, infections, clotting, bleeding problems – that could be mimicking or causing these symptoms.” And suddenly, he says, he realized that “what people had been concluding about shaken baby syndrome, they were assuming.”

Barnes, who is now at Stanford university, was not the only one undergoing a change of heart. Today, he is part of a small group of doctors who travel the country as witnesses and argue in journals that the shaken baby diagnosis needs an overhaul. Their names alone provoke hostility among many pediatricians, most of whom support the diagnosis. “These people are often pathologists who don’t have any clinical experience,” says Reece. “They see the dead tissue at the end of the process, and they have no information other than what they read and hear from colleagues. They’re speaking out of turn.” Yet many of the dissenters are well-credentialed doctors – and their work has shaped medical knowledge of the syndrome, even if only by challenging mainstream doctors to defend their ideas. Dissenters argue that:

A person can’t shake a baby hard enough to cause the cardinal symptoms of SBS. Woodward’s defense team argued that she could not have shaken a baby as heavy as 25-pound Matthew with enough force to cause his injuries. The argument remains common in court and relies largely on biomechanical experiments. Dr. Jan Leestma, a Chicago forensic neuropathologist who has been one of the most visible SBS defense witnesses, says those experiments (and what he calls “the principles of physics”) were important in fueling his skepticism of the diagnosis. “When the biomechanics papers started coming out, I thought, ‘This is deep doo-doo,’ ” says Leestma. But Reece, who says the point “is specious,” contends that it’s impossible to measure how much force could kill a baby: “It’s not like we can say, ‘It takes 17 shakes at X many G-forces.’ That’s a fool’s errand.”

SBS symptoms can result when the baby is struck against a hard surface, but shaking is not enough. In the summer of 2005, courts in the United Kingdom reexamined several shaken baby cases after British neuropathologist Dr. Jennian Geddes traced most child-abuse victims’ brain injuries to sudden, blunt impacts, not shaking. Whether shaking alone can mirror the effects of blunt trauma continues to be widely debated. Cases that clearly involve blunt trauma are now called “shaken impact syndrome.”

The symptoms do not always manifest immediately after shaking, so the time of shaking cannot be pinpointed. Dr. John Plunkett, a Minnesota pathologist, wrote shortly after the Woodward trial that the symptoms of SBS might begin long after the initial injury. Other doctors have reported caring for shaken infants who experienced an extended “lucid interval” before succumbing. This argument is popular in court, because it throws doubt on who might be the perpetrator of the child abuse. But Reece, like most doctors, says the toxic cascade in the brain begins “very, very shortly” after the shaking, usually within minutes. “I think the claim that these things happen days or even weeks before is contrary to everything we understand in pediatric clinical medicine,” he says. “You can’t have a serious injury to a baby’s brain happening even a day before the child deteriorates and goes into a coma. It doesn’t work that way.”

The symptoms of SBS can also be caused by falls from short distances. Plunkett has provided ammunition for this theory. In a 2001 paper, he described a girl, not quite 2, who died after falling off a 2-foot-high plastic jungle gym. Investigators accused her parents of shaking her – until they saw a videotape of the fatal fall. Biomechanics experts continue to examine whether trauma from short falls can cause injuries like those diagnosed as SBS.

All of these arguments have played out against the background of “evidence-based” medicine, the movement to incorporate proven data as well as clinical tradition and individual observations into medical practices. Barnes argues that he and others are trying to bring that sort of scientific rigor to a field sorely lacking it. “The diagnosis and management of child abuse has escaped evidence-based medicine and a good methodology for 30 or 40 years,” he says. “It’s supposed to be founded in the scientific method and biostatistical analysis.”

But Block calls this argument a “smoke screen.” Most evidence-based medicine, he argues, is founded on randomized trials, while child abuse studies can never be. “The main thing we don’t know is what happens inside a real brain, a real head – and we will never know,” he says. “How are you going to find out? Are you going to take 100 babies and shake half of them and see what happens?”

Underneath the sharp words, both sides say they want the same thing: a solid diagnosis, the kind that will not be debatable in court. Therein lies the heart of the problem. When no one confesses to shaking a baby or witnesses the crime, doctors are often the only ones who can provide anything close to definitive insight. But in some shaken baby cases, depending on the injuries, the doctors’ detective work simply cannot be definitive. It’s a case of conflicting standards: Criminal justice calls for proof beyond a reasonable doubt, and medicine alone can’t always provide it.

This mismatch is one of the few points on which both sides can agree. Reece, the pediatrician, says shaken baby syndrome is unique in its differing medical and legal standards of evidence – and he doesn’t know how to reconcile the two. “You can’t be black and white in medicine, and that’s the problem, because when it comes to court, you have to be black and white,” he says. “Should people be convicted on less than 100 percent foolproof evidence? There’s not always an absolute airtight case. I don’t know what you can do about that.”

For Leestma, the forensic neuropathologist, the discrepancy leads to an inescapable conclusion. “There can’t be a relaxed standard for one branch and not the other,” he says. “We tend to be a little sloppy in medicine with diagnosis and treatment, but the law demands a more rigorous approach. If we don’t know what happened, it ends there.” Blake, the defense lawyer, is even more certain. She argues that in the rare cases that rely on medical evidence alone, that evidence “should not equate to ‘beyond a reasonable doubt.’ To me, that should not even pass probable cause.”

Some courts are starting to lean toward Blake’s position. Judges in Kentucky and Florida recently banned the term “shaken baby syndrome,” and medical testimony has gotten a number of convictions overturned. In addition to the British cases, there have been several in America, most notably that of Ken Marsh, a San Diego man who was convicted in 1983 of killing his girlfriend’s young son – and exonerated in 2004 after an appeals judge concluded the child’s injuries came from falling off a sofa. Blake, who is close to the family, officiated at Marsh’s wedding to the girlfriend who stood by him during his 21 years in prison. Marsh is also suing for wrongful prosecution, with Blake’s help; experts will be chosen next month.

Another case, making its way through the Wisconsin appeals system, eerily resembles that of Louise Woodward’s. Audrey Edmunds, a baby sitter imprisoned for almost a decade for fatally shaking a 7-month-old girl in her care, will have her second set of hearings this month. Her lawyers are expected to argue that the science used to convict her is now invalid and are tapping as the star witness Robert Huntington, a forensic pathologist who originally testified against her. He “now unequivocally rejects his prior opinion and its implications and will testify to that effect,” according to a brief filed on Edmunds’s behalf. He cites the “lucid interval” of the victim – or what Block skeptically calls the “somebody-else-did-it theory.”

Although there are no prominent appeals on scientific grounds in Massachusetts, shaken baby cases continue to be a problem in the state. From 2001 to 2004, almost 80 Massachusetts children were diagnosed with SBS. The days just before Christmas saw a horrifying potential case: 9-week-old twins from Dracut were rushed to the hospital with injuries consistent with shaking. One of the children, Paige Garvey, is now in foster care. Her brother, Liam, died at the Floating Hospital for Children on December 22. At press time, no charges had been filed.

For Jetta Bernier, the Dracut twins case was especially heartbreaking. Just weeks earlier, thanks in part to her lobbying efforts as director of Massachusetts Citizens for Children, the state had passed what she calls “the most comprehensive legislation in the country” regarding shaken baby syndrome. Late last year, Governor Mitt Romney signed a law calling for a statewide prevention plan intended to help researchers gather data and, says Bernier, provide training for new parents on how to cope with ceaseless crying. It’s unclear, however, whether the law will actually translate into action. Trying to raise funds for a similar plan several years ago, Bernier says potential donors told her there was no need, since the Woodward case had already publicized the dangers of shaking. A few others told her the defense had persuaded them the syndrome didn’t exist.

Now Bernier finds herself struggling for funding again. There is no appropriations bill attached to the law. “At this point, there’s no funding at the [Massachusetts] Department of Public Health to carry this out,” she says. “We and others have been able to pilot efforts in a couple of counties, but there’s certainly not enough money to take this full scale.”

Another person who watched the Dracut case with dismay was Paula Cashin, a longtime resident of the town. Well before Louise Woodward ever met the Eappens, Cashin was a sad and unwilling authority on shaken babies. Her 7-month-old daughter Laura began to show signs of SBS on August 30, 1990, while in the care of a baby sitter who lived down the road. That night, at the hospital, Cashin and her husband sat through separate interviews with the police while their daughter spiraled downward. By morning, Laura was brain-dead. It was then, Cashin says, that the police told her the baby sitter had “admitted to a state trooper to yanking Laura out of her high chair by one arm, and demonstrated how violently she shook her.” But in court, the baby sitter admitted only to shaking the baby gently and said she had never intended harm. She was found not guilty of manslaughter in 1992. “At least in the Eappen case they found [Woodward] responsible,” says Cashin, who was one of many supporters reaching out to the Eappen family during the Woodward trial. “All I ever wanted was for that woman to tell me, ‘I did this, and I’m sorry.’ And I will never know that.”

The Eappens can understand the sentiment. Asked what an apology from Woodward would mean to her at this point, Deborah Eappen says she thinks “there’s a lot to be said for saying you’re sorry.” But, she adds, “I’m not holding my breath.” She does take some comfort in the original verdict of guilty of murder in the second degree against Woodward, despite the judge’s later reduction of the charge to involuntary manslaughter. “The really important thing,” she says, “is that we know the truth, and the jury saw it.”

The other important thing to the Eappens now is preventing more cases. The family’s Matty Eappen Foundation cosponsors national conferences on shaken baby syndrome and has helped put together diagnosis-training workshops for physicians. It also recently started a scholarship for the siblings of shaken babies. As for Matthew’s own siblings, who are now 12, 8, and 6, they regularly field questions about their brother at school. “It just ripples through your life,” Deborah Eappen says. “I feel like we can’t run away from this. This is where we are. It’s not like our kids have a normal life, but we try.”

Deborah Eappen does not keep tabs on Louise Woodward, but occasionally, she says, acquaintances will mention news reports to her. The last one she heard about was when Woodward famously began a career in law in 2004, declaring it a “crusade” inspired by her trial. It was not a long one, though; little more than a year later, Woodward left her firm to teach salsa dancing. Woodward did not respond to an interview request and has not commented publicly on any of the research from the last decade that she put her confidence in 10 years ago – or whether, as she hoped, it has bolstered her case.

Gerard Leone Jr., the current Middlesex County district attorney, who helped prosecute Woodward, has a firm opinion on that matter: “I think there’s been a lot of good scientific work to try and enlighten us, but I don’t think any of it has shown in any way that what happened in that case was anything but child abuse.” Reece, the pediatrician – who was scheduled to testify against Woodward and was dismissed at the last minute – also says there’s nothing scientists have learned that would change his mind about the verdict. But, perhaps unsurprisingly, Leestma, the forensic neuropathologist who often works with the defense in SBS cases, is not so sure. And Barnes, the radiologist who changed sides on the syndrome, flat-out disagrees. “The science [we have today] could, in fact, have exonerated her,” he says. “There is certainly, in retrospect, reasonable doubt.”

This, to Eappen, is precisely the defense mechanism that “needs to be overcome,” a “denial that takes in place in the courtroom by everyone who tries to think of another excuse for what happens.” She can understand, she says, the reason for that denial: “It’s too hard to imagine someone doing this to such an impossibly defenseless victim. You always wish it’s something else.” Cashin, who lost her daughter, has a similar take, but she describes herself as more “jaded.” Once a case reaches the courtroom, she says, “the truth really doesn’t matter. There are so many ways around it.”

She may be right, but in a much less jaded sense than she intends. In the courtroom, after all, it is the jury that makes the decision. Ultimately, whether the doctors arguing over shaken baby syndrome can state their cases beyond a reasonable doubt may be irrelevant. The question is whether the jurors, hearing all of the evidence, can. In the Garvey case, in Wisconsin, and in courtrooms across the nation, they, and not the doctors, must come to a consensus. Justice is a matter of their opinion, and that’s one thing since the Woodward trial that hasn’t