Does the Brain Injury Prove Abuse, or Not?

The two shaking convictions in the news this season, ironically, are both based on brain findings alone, with no bruising, bone injuries, or other signs of assault—at least five years after the experts started saying that they never diagnose shaken baby based on the triad alone. In the widely reported case of Joshua Burns in Michigan, in fact, the diagnosis was based on only two of the three markers, retinal hemorrhages and subdural  hematoma.

A commercial airline pilot active in his local church and community, Joshua Burns was convicted of child abuse in December of 2014. Supporters immediately launched a web site and media campaign protesting his innocence, and now the Michigan Innocence Clinic is appealing his conviction.  Naomi Burns, now 18 months old, seems to have recovered fully. Her mother Brenda Burns has regained custody, but Joshua has been denied any contact with his daughter since April of 2014.

In New Jersey, meanwhile, child care provider Michelle Heale has been sentenced to 15 years in prison, with the requirement that she serve at least 85% of her sentence. She was convicted of aggravated manslaughter and child endangerment in April.

Michelle said 14-month-old Mason Hess choked while eating a tube of applesauce, but doctors concluded from the brain findings that he had been shaken to death. The only visible evidence of trauma was a bruise on his forehead, suffered several days earlier in a fall in his home. He had been sick with coughing and copious mucous, and had just started taking medication the day before.

The coverage of the sentencing last week brings home how an abuse diagnosis amplifies the pain for everyone. The child’s mother asked the judge to impose the maximum sentence for “the monster who stole our son’s life,” and her husband addressed Michelle directly:

We truly believe that you should die for what you did to Mason, just not yet. But when you do, your true sentence will begin. We know that the devil himself has saved a special place in hell for you because only the most evil being would hurt an innocent child, an innocent child who cannot protect himself.

Judge Francis J. Vernoia imposed only half the maximum sentence of 30 years, but he chastised Heale for refusing to “take responsibility” for Mason’s death. Kathleen Hopkins at the Asbury Park Press wrote:

Vernoia noted, however, that Heale is at risk of committing another offense because she has refused to take responsibility for her actions. The judge said that despite overwhelming evidence of her guilt, Heale told a doctor during a mental examination that her conscience is clear.

The judge saw overwhelming evidence of her guilt, but I don’t.

The prosecuting attorney told the court that the child suffered “severe brain, spine and retinal damage,” but the injury to the spine was microscopic blood identified at autopsy, a finding consistent with the time the boy spent on a respirator. In fact, the prosecution was based on subdural hematoma, retinal hemorrhage, and brain swelling.

The historical context:  In 1998, after Judge Hiller Zobel released convicted British au pair Louise Woodward with a sentence of time served, several dozen child abuse experts signed a letter to the journal Pediatrics protesting media coverage sympathetic to the defense. The letter made this statement about the diagnosis of infant shaking:

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—swelling of the brain (cerebral edema) secondary to severe brain injury, bleeding within the head (subdural  hemorrhage), and bleeding in the interior linings of the eyes (retinal hemorrhages). Let those who would challenge the specificity of these diagnostic features first do so in the peer-reviewed literature, before speculating on other causes in court.

Whether or not you accept that guideline—which I don’t—the prosecution of Michelle Heale is consistent with it.

In about 2010, however, after the Audrey Edmunds reversal, I started hearing a different message at child abuse conferences.

At the Eleventh International Conference on SBS/AHT, for example, Dr. Robert Block, then the incoming president of the American Academy of Pediatrics, rejected “the so-called triad,” with the comment:

Only people who are not active physicians working with children, naïve journalists, and professors with a biased agenda would propose that only three signs and symptoms support a diagnosis.

And in 2011, Dr. Carole Jenny presented at the New York Abusive Head Trauma/Shaken Baby Syndrome Training Conference, using a PowerPoint that featured this statement:

No trained pediatrician thinks that subdural hemorrhage, retinal hemorrhage and encephalopathy equals abuse. The “triad” is a myth!

But the conviction of Michelle Heale seems to be based on the presence of the triad, combined with faith that the effects of a serious shaking assault are always immediate. My best guess is that the diagnosis rested largely on the extensive bleeding in the child’s retinas, bleeding that the child abuse experts insist is caused only by abuse. During the trial, pediatric ophthalmologist Dr. Alex Levin was quoted in USA Today as saying, “This is a victim of abusive head trauma. I can’t think of another explanation.”

February 2022 Update:  Law Professor Colin Miller in New Jersey has filed an application with the state Conviction Review Unit, requesting that they “correct an injustice and set Michelle Heale free.”

Retinal hemorrhages were also behind the diagnosis in the Joshua Burns case, where the eye exam did not take place until 11 days after the presumed shaking. The Torn Family web site, about the Burns case, offers both prosecution and defense expert opinions, making it possible to unravel some of Naomi’s complex medical history.

Her delivery was prolonged and difficult. Four times, the doctors attached the vacuum-extractor to Naomi’s head and pulled, and four times, the cap popped off. Finally, she was delivered by C-section.

She had difficulty latching onto the breast, and she was gaining weight slowly, so Brenda was pumping breast milk and feeding her daughter with a bottle, which was easier for Naomi to suck on. The girl’s  head circumference was growing at a faster rate than the rest of her body.

Then one Saturday afternoon when Naomi was about 11 weeks old, Joshua was at home alone with her, holding her on his lap. Brenda had gone to the hair salon, where the computer asked for a PIN number to go with the credit card. She called her husband, who gave her the number and then, he says, while putting the phone down, almost dropped Naomi, catching her by the head and face before she hit the coffee table.

When Brenda got home, Joshua told her what had happened and showed her the red mark on Naomi’s face. Brenda says she thought it looked like a fingernail scratch, something she’d seen many times in her years as a nurse. (She had quit her job at a local hospital just a few months earlier, in preparation for Naomi’s birth.)

Their daughter seemed fine that night, Joshua and Brenda say, but over the next several days they took her again and again to the doctor and hospital for vomiting, pallor, and uneven breathing, only to be reassured and sent home. On the third day, when she had a seizure during an examination, she was admitted to the hospital. The first reading of the intake MRI identified only “benign enlargement of the subarachnoid spaces” (BESS), which means there was a little more room than usual between the brain and the skull, a not-uncommon condition of infancy that usually resolves on its own. It does, however, predispose to subdural hematoma.

Doctors attributed the infant’s seizures to an unidentified gastro-intestinal virus, and after a week in the hospital she was sent home, with anti-seizure medication. Her parents brought her back within hours, though, worried about more vomiting, fussiness, and pallor. At that point, while following up on a possible metabolic disorder, doctors performed a dilated eye exam and were surprised to find what the child abuse report identified as “bilateral multilayer retinal hemorrhages extending to the periphery,” the pattern considered by most child abuse experts to result exclusively from shaking.

“The retinal hemorrhages were an incidental finding,” Brenda says, “but they changed everything.”

Radiologists re-examined the initial MRI and reached a new conclusion:  They still determined that Naomi had BESS, but they also saw old subdural bleeding, possibly dating back to her traumatic birth, and a small amount of fresh subdural blood. Naomi was diagnosed as a shaken baby, and prosecutors targeted Joshua as the abuser.

The defense brought in pediatric ophthalmologist Dr. Khaled A. Tawansy, a retina specialist, who wrote in his report:

The retinal hemorrhages that were seen by the ophthalmologist at University of Michigan and documented by Ret-Cam imaging were predominantly superficial (sub internal limiting membrane or nerve fiber layer or intra-retinal)… [These types of hemorrhages] occur regularly with abrupt elevations of intra-cranial pressure (as in acute subdural hematoma) when the pressure in the cerebrospinal fluid surrounding the optic nerve exceeds the pressure of venous return in the retina as it drains into the optic nerve.

The child abuse doctor in the Burns case asked for advice from the same pediatric ophthalmologist who testified against Michelle Heale, Dr. Alex Levin in Philadelphia. In an email exchange posted on the Torn Family web site, the child abuse doctor summarizes Naomi’s medical history, acknowledging but discounting the traumatic birth. She reports the eye findings in detail and notes that the child has “persistent thrombophilia,” a clotting disorder.

Dr. Levin’s response:

Impressive documentation. Very well done.

Not sure what the question is. I can’t think of another diagnosis other than abuse assuming no obvious coagulopathy or other event.

Apparently Dr. Levin missed the mention of thrombophilia in the original note, because later in the thread, when the pediatrician asks him about it, he writes:

Do you mean thrombocytosis?
Either way we have no idea what this might do re retinal bleeding and could be considered to throw the retinal findings into question. We just don’t know

I don’t know what was said about all of this in court, but it seems to me that retinal hemorrhages are starting to play a very large part in these diagnoses. In this case, the retinal hemorrhages themselves are suspect, and the only other evidence of abuse is subdural hematoma in a child with BESS.

Now the state is trying to terminate Joshua’s parental rights. Those hearings wrapped up in June, and the family is waiting for the judge’s decision.

I don’t understand why the child abuse doctors are so sure Naomi was abused, and I’m even less clear on why the state has decided that her father and not her mother must be guilty. The accident Joshua reported—presumed to be a lie invented to cover up abuse—happened 11 days before doctors ever considered the possibility of head trauma.

I am sorry that another family has been caught up in this tragedy, but I am heartened by the enthusiastic support that Joshua and Brenda Burns are receiving from their community. And I have high hopes that the ongoing coverage will bring more light to the debate about the reliability of a shaken baby diagnosis.

September 2015 update: The court has adopted a treatment plan that allows for reunification of the Burns familly, which is good news if you ignore the irony that the only thing between  the Burns family and reunification is the court.

copyright 2015, Sue Luttner

If you are not familiar with the debate surrounding shaken baby syndrome, please see the home page of this blog.