Interview Dr. Michael Laposata

He is the pathologist in chief at Vanderbilt University and specializes in patients with bleeding and clotting disorders. Laposata says such disorders can manifest in ways that look like child abuse. This is the edited transcript of an interview conducted on April 27, 2011.

• Why he's predisposed to look for blood disorders if bruising is the only evidence of abuse
• "We're underdiagnosing child abuse far more than we're overdiagnosing it, but what happens when we overdiagnose it?"
• "Not every case is conclusively diagnosed at autopsy…"
• How many bleeding disorders are there that can mimic child abuse?

You're an expert in both pathology and blood-clotting diseases?

Yeah. Pathology is the broad spectrum. It is the study of the diseases. ... So I suppose what you could say is I know how the diseases develop, and I also know how to treat the diseases. That's the medical part. ...

How is it that you came to be someone who looks at possible child abuse cases?

One of the major manifestations of child abuse is a bruise. ... So why couldn't a child who may have been abused, who has only bruising as a finding, simply have a bleeding disorder? That's how we got involved.

And as you started looking at these cases where parents were accused of abusing children, what did you find?

The first case in which I got involved was a case in which there were clear signs of bruising and bleeding before the event that was thought to be associated with child abuse actually occurred. And because of those signs of bruising in advance, it raised the question that this child might have been born with a bleeding disorder and that just a mild injury could produce a big bruise. So that was probably the clue in the first case.

In the second case, there was no obvious bleeding or bruising ahead of time, but there was an awareness that you could go through your early days without having enough bumps and bruises to identify you as a bleeder. So we simply did the blood tests, ... and sure enough, we found a bleeding disorder in three children, one whose father was accused of child abuse and then his two siblings.

And then other cases appeared where there were clear abnormalities in both lab tests and easy bruisability. And it's not always just easy bruising. In one family in which there were four sibs, three of who bled, they would not only have easy bruising, they also had significant nosebleeds. The mother told me the story one night that they all sat down to dinner, they said grace, and then three kids started to bleed. And they were all the kids who had the abnormalities in the clotting factors.

So you've got these cases where children are coming to you. They're bruising; they're bleeding. Their parents are suspected of abusing them. ... What do you do? What is your plan to figure out what's happening?

So the first step is to find out if there's any evidence of a bleeding disorder. And I'm predisposed to look for it if all we have is bruising, because if the child has burns on the skin, if there are broken bones, old fractures, it's much more likely that there's child abuse. But if it's just bruising, that puts it into a different category.

So we start with the simple tests, the tests that everybody does to look for a bleeding problem. ... I think that the important thing to understand as we go forward is you can't stop with the simple tests. They are often known as the PT [prothrombin time], the PTT [partial thromboplastin time], and the platelet count. But you have to look for things like von Willebrand disease, which would not be made clear by doing those other tests. You would have to specifically look for that. ...

The challenge in all of this is sometimes the patients are really small. How much blood can you get from a living child who [is], say, three or six months old? Because these blood tests might require small amounts of blood, but in a child it's a large amount of blood. So you can do the initial tests, and then you may have to come back and do further tests. You can't reach your conclusion after the first analysis often because you've only done the simple tests. ...

And when it comes to small children, what is the concern if you don't properly diagnose a bleeding disorder? What happens then?

In these cases of is it child abuse, you have the potential for a misdiagnosis. Now, what's very clear is that the major misdiagnosis out there with child abuse is missing it. So what happens is this child is abused, but for whatever reason, they go back into the home where they're being abused, and when you miss that diagnosis, that's terrible, obviously.

So we're underdiagnosing child abuse far more than we're overdiagnosing it, but what happens when we overdiagnose it?

Now, when you have the outrage of seeing the child who is lifeless and beaten and your first reaction is, "We have to find the person who did this," you have to stop for a minute. You have to step back, take an analytical look. ... We have to figure out if this really is child abuse, because overdiagnosis may be even worse than underdiagnosis, because not only are we affecting the child, we are now affecting another person, a parent who is being accused of abusing the child.

You've looked at a lot of these cases where parents are suspected of abusing their children. What happens if you get it wrong? What are the consequences?

Terrible. So in the first consequence, in the first example, this young man who was accused of abusing his daughter, who clearly had von Willebrand disease, was given the opportunity at sentencing by the judge to either admit that he had hurt his daughter purposely or to go to jail. And he said, "I love my daughter, and I simply can't say that I did it, because I didn't." They put the handcuffs on him and took him to jail.

In the second case, the father had to live in the neighbor's house because he was not allowed to be unsupervised with his children. So for an entire year, they were in this unusual situation. And the mother worked into the evening hours, and he had always given the kids their supper, gotten them ready for bed -- all of that had to stop. So they no longer had their father in the traditional way that they had had him because he was not permitted to be unsupervised with his children. ...

In the case of the woman with the three out of four bleeding children, she said she was accused of abusing the children on multiple occasions, [once when] they were standing in front of the supermarket where somebody saw three bruised children with her and reported her to the police, and they came to her house. So she never went to jail, but the consequences of these problems are immense.

You told me that people who don't understand blood-clotting disorders don't understand some of the things that you've seen. ... Tell me about the kind of symptoms that people can have when they have these kind of disorders and diseases.

It's surprising actually, but one of the areas that many physicians understand the least is blood clotting. ... About 30 years ago, the field of hematology merged with the field of oncology, and they created a discipline called hematology-oncology. And when they created that discipline, they decided to pay for cancer care far more than caring for hematologic problems, so people who entered this discipline mostly became cancer doctors. And therefore there was not that same sense of urgency to understand the hematology, especially the clotting, and that, over the past 20 or 30 years, has reduced the number of people who are experts in coagulation.

We've been looking at cases where people were wrongly convicted of killing children or possibly wrongly convicted. We went to Canada; we interviewed Justice [Stephen] Goudge, who oversaw an inquiry into people being wrongly convicted of killing children there. One of his recommendations was when a forensic pathologist is dealing with a child death case, they should consult with other experts to make sure that they understand all the possible diagnoses. Is it something that you think is a good idea?

Oh, it's a great idea. It probably doesn't happen much. I think that if you look at many of the cases that appear in the forensic pathology suite, they are often straightforward -- a gunshot wound, a stab wound. But I think it's important to understand that not every case is conclusively diagnosed at autopsy, and bleeding disorders is one of them. ...

The diagnostic testing to find out if you have bleeding disorder requires a live patient and flowing blood, so you see, this is one of those cases where to identify a bleeding disorder, you can't do it in the forensic pathologist's office; you have to do that in a hospital. ...

And in your experience, how often do the worlds of forensic pathology and the worlds of academia and medical research and clinical treatment really overlap?

Not enough. I think that there probably are times when they overlap, and it would be highly dependent upon the forensic pathologist who would look at the case and say: "I need help. Why don't I call somebody?"

And I'll bet you that there are some very special forensic pathologists who do this and do a great job because of that. On the other hand -- I certainly have seen it with this child abuse -- is that the personal outrage of the circumstance stops people from being analytical, and then they stop asking questions.

And what happens?

And then a misdiagnosis is given a chance to occur.

And in the end?

And then somebody goes to prison who shouldn't go to prison. ...

You have a PowerPoint presentation that compares children with bleeding disorders to children who definitely have been abused.

Right.

What happens when you show it to people?

So I usually tell people in advance that the pictures are graphic because these are children who are bruised. But what I did was I took pictures of children who had a bleeding disorder and I put it right next to a picture on the same slide of a person, a child, who suffered child abuse.

And the title of the slide is "Which one of these cases is child abuse?" And nobody can ever tell. ...

And what do you think of that?

Well, it shows you the magnitude of the problem. ... I've been looking at patients with bleeding problems for years, more than two decades. And if you show me the two children with the bruises on their legs, I couldn't tell you that that one is the bleeding disorder. I'd have to do the blood test to find out.

We've been looking at cases where children died, and medical examiners doing autopsies of these children either didn't consult the medical histories and the medical records of the children or didn't consult them very closely. What do you think of that?

Well, gee, just imagine if you came in, and you had the worst headache of your life. And we just said, we're not worried about any of your blood tests; we're not going to do any imaging studies. I'm just looking at you, and since it's the worst headache in your life, I'm going to presume it's a cerebral aneurysm, so I'm going to make an incision right here and look at the bottom of your brain.

I mean, come on, right? You have to put the whole picture together to be able to get to a diagnosis. It's a 500-piece puzzle. And sometimes that 500-piece puzzle is a snowstorm, and it takes a long time to put the pieces together.

So in those circumstances, it seems overly presumptuous that you could just by looking at something figure out what it is. There are lots of other diagnostic pieces usually within your grasp that you need to pursue, understand, and fit them into this big puzzle until you have some satisfaction that's the reason. ...

So let's talk about Isis [Vas] in a little more detail. This is a six-month-old baby who comes to the hospital with marks on her body, and she's allegedly bleeding from her vagina. She has dark, tarry stools. And the police arrest the man that last had custody of her, Ernie Lopez. When you looked at her lab reports, you looked at her medical records, what made you think the police might have gotten it wrong here?

If you look at some of the findings in the lab studies that were done, and in the presentation, some of them take longer to evolve than the time period during which Ernie and the child were together.

So dark, tarry stools don't appear in 40 minutes. So if you look at how long it takes to elevate your white count to get dark, tarry stools, to turn your liver function tests abnormal, something had to be going on for days, days. ...

... The medical examiner who did Isis Vas' autopsy [was asked], "Is it customary for you to look at lab reports before doing the autopsy?" And the medical examiner replied, "A lot of times we don't have the lab reports." She went on to say that basically she didn't always consult lab reports or medical records when doing autopsies or drawing her conclusions. What do you think?

So I think part of the problem is that there is an opportunity for better communication among doctors. Is it quite possible that patients appear and they don't have the lab tests? Absolutely. And it could be that after you've gotten used to not having the lab tests, you don't think about what you would do if you had them.

But the reality is, especially for a disorder which you can only diagnose in the living patient, like a bleeding disorder, you have to look at the lab tests, because it's your only clue. The bruise on the expired patient tells you nothing.

So in the case of Isis Vas, she came into the hospital and had a whole battery of tests done on her before she passed on.

Right.

And that could have given the medical examiner clues that she didn't look at.

There were clear abnormalities even in the routine lab tests in this case. The tests called PT and PTT were markedly abnormal. The platelet count was low. There's another test called fibrinogen that was low. It was a classic picture as you're putting in this puzzle together of a disorder called DIC, disseminated intravascular coagulation, and you can bleed from that.

And there are plenty of different causes, and one of them is infection. So if you look for causes of infection, there were some clues in this -- an elevated white blood cell count; she'd also had some bleeding; she had a tarry stool, so there was blood in her stool. So it was clear that she'd been bleeding for some period of time. And she had lab tests associated with the bleeding disorder. So it was a pretty good speculation that there's an infection as a cause of this, but the one thing that's for certain is there is a bleeding disorder.

Now, that said, one could not rule out the possibility that a child with a bleeding disorder is abused. So I think that probably makes a lot of people uncomfortable, especially in the legal system, where somebody could say, "Well, I don't care if there was von Willebrand's disease; this patient was beaten by the father." Well, I think if the question in the legal system is do we have some level of doubt based upon the presence of a bleeding disorder and that's what we're looking for, there's clear doubt here, because this child had a significant bleeding disorder.

When you reviewed the laboratory results for Isis, what were you thinking? How did these results compare to other patients that you've had?

Sometimes when you look at a report, you only have a few tests, and maybe there's a minor abnormality. It's very hard to interpret those because it's hard to know if that minor abnormality really could have produced the kind of bleeding that was observed.

The difference in this case is that the abnormalities were significant. They were so very abnormal, it wasn't a question of could this be associated with increased bleeding? It was obvious, so there was not a question in this case. And some of them are more challenging.

So you look at these lab reports, and you say, "This is a child with a very serious bleeding disorder"?

Correct.

I'll read you something else that the medical examiner in this matter said. She was asked about the blood-clotting test, and she said: "I don't get into PTT. I'm a forensic pathologist, and all my people are dead. We don't run PTTs." ...

You obviously highlighted the point that was made earlier, that many people don't understand coagulation well enough. ...

Clearly, if we're talking about bruising, the presence of a prolonged PTT and abnormal PTT test is related to that. So I think you have to know. You just have to say for the benefit of my patient, for the benefit of getting the right diagnosis every time, I have to get into it.

From the time Isis was brought into the hospital, she was treated as a suspected child abuse case. You noted when you reviewed the case that tests that would have been really helpful were not run on Isis. What are those tests, and what can help diagnose child abuse or a bleeding disorder before a child dies?

When you have a prolongation of the PT or the PTT, there are many further studies that you could do to find out why. ... There are tests that you can use if you have the living patient to follow up to not only say general-category bleeding disorder, but you have DIC; you have vitamin K deficiency; you have DIC because of leukemia. And you can put all the pieces of the puzzle together. So that's when it becomes much more comfortable to say we know there's a bleeding disorder. ...

The state's experts in this case, they'll say: "Dr. Laposata, you got it wrong. When a child or an adult has head trauma, they can get a blood-clotting disorder. So you know that must be what happened, that the child was beaten and got this clotting disorder. You got it wrong." What do you say to them?

Well, they got it half right. I think it's fair to say that there are other stimuli for what we're calling DIC. One is infection, which is what we have proposed based upon the findings. Another one is trauma, so, yes, you could be hit on the head; yes, you could have an auto accident, and you could go into DIC.

The trouble is that all those findings that we're talking about -- liver function; test abnormalities, in particular the tarry stools -- if she were hit on the head, they don't occur in the short number of minutes that we're talking about.

So there are multiple facts to consider here, and a lot of them relate to the temporal association of events. So if you just put the pieces together, you have to remember that you're putting together a puzzle for the patient at time zero. And then 10 minutes later it's a different puzzle, because now it's evolving. And you could easily make a case when something is happening for days, like an infection that this occurred. But I think it's a pretty difficult case to make that the inciting event for the DIC was trauma, and that over some short number of minutes all these abnormalities appeared.

Because the lab results were just so abnormal that you wouldn't expect them to crop up in a really short period of time.

In minutes. In minutes.

In your practice, that's not something that you've seen a lot of, I take it?

Correct. Correct.

So in the case of Isis, who died -- and this is a case you reviewed -- she had vaginal bleeding, and that was taken by the doctors who treated her and by the medical examiner who examined her as a sure sign that this child had to be sexually abused, that this was the only explanation for it. Is it possible that there is some other explanation?

Oh, sure. I've had a patient once who unfortunately lost the baby in the ninth month. She had a fetal death in utero. And that is also another cause for DIC. When I saw this patient, she was bleeding from every orifice, every orifice. So there was vaginal bleeding; there was bleeding in the mouth; there was bleeding from the nose. So when you have rampant DIC you can just bleed from everywhere. So I think that to conclude that [Isis] was sexually abused is an overstatement given what we know about the whole picture in this case.

And when you see these lab results, you say this is rampant DIC?

Yeah, absolutely. Absolutely.

It's not a mild case?

It was not a mild case.

You co-authored an article about diseases and disorders that can mimic child abuse symptoms. Tell me about that.

The reason that article was written is because when I testified in the first case, the prosecuting attorney said, "How many cases are there in the medical literature that have been reported that show kids with [diseases] that look like child abuse?"

And it turns out there were really none. There was one that was a maybe, but then I realized that the problem is that doctors don't write up and publish all of what they see. In fact, it's only the tiniest percentage.

So for the last 20-some years, I've been reviewing 10 or 20 cases a day of people with bleeding or clotting disorders. And so every day I see one or two that are unusual presentations. There are new [pieces of] information, but they don't appear in the literature, because if all I did was to write the cases, I would have no time to do all of my other responsibilities. ...

I thought it was appropriate to collect all of what was out there. So we did an entire search of the literature, and we found children who had vitamin K deficiency, who had bleeding. We found people who had ITP; they had the flu; their platelet count went down; they had idiopathic thrombocytopenic purpura (ITP). ...

So that article is a compilation of all of those, and since that time I've been able to reference that article when people say, "So, what other things masquerade as child abuse?" And now we have at least got a collection of things that are published. ...

How many major diseases or disorders do you think mimic child abuse symptoms? How many are out there that are prominent?

I think that there's probably at least a dozen prominent bleeding disorders, and then there's plenty hundreds that are not so prominent. The question is how many of these people, how many of these kids with a bruise -- just a bruise now; no burns, no broken bones on top, just a bruise -- how many of these could potentially be misdiagnosed?

I think it's a scary number. ... What if 1 to 2 percent of those are misdiagnoses? We probably have hundreds of men in jail because their child had a bleeding disorder. They had a minor injury, and the father was disbelieved. And so I think the ability to go back and say, "What were those findings, and let's review them again," much like when DNA evidence appeared, where people were re-evaluated and it was found, goodness, you know, the DNA evidence shows that it was somebody else, and it was not you.

I think it's time to look at these cases again and to say is there some doubt, was there really a coagulopathy here, even if we look at the simple test that I think would result in the liberation of a number of falsely accused people. And I don't think it's a handful. I think it's far more, and probably in the hundreds.

That is a haunting notion.

It's a haunting notion. I will tell you that when I was holding my six-month-old daughter in the middle of this first case of child abuse, and at six months, with her wiggling, she was sitting on my lap at the table, and she got floppy, and she went bunk, right on the table with her head.

And I thought, if she has von Willebrand's disease I'm going to jail. And that was the scary thought about it, because who is going to believe me? ...

We've been going around the U.S. and Canada, and what we hear from forensic pathologists is that a lot of times the mentality in the morgue is if you see a dead child, suspect immediately that the child was murdered, that this was child abuse. Is there a problem with that?

Well, yes, of course. I can understand how that thinking occurs. ... But I think what's wrong with it is there is an acceptance that we could get it wrong by overdiagnosis, and that's OK. That part is not OK. ...

And the medical examiner who did the autopsy [on Isis Vas] she says: "Dr. Laposata, you're wrong. This child had marks from bruises where someone must have struck her. She had bleeding from her vagina, so she must have been violated. It had to be trauma because of these marks, because of this bleeding. And she had bleeding on her brain." What do you say to that?

... I certainly know that if you have trauma, you can produce a bruise. But the reality is, when your blood is so thin, when you're so unable to make a clot, you can just develop bruises, and they can be spontaneous. So just because there were bruises doesn't mean that the child was ever hit.

In this case, there's a report that the child developed a bruise after being strapped into a seat belt, a bruise up by her neck or shoulder area. Does that raise a red flag for you?

It does, because think about it a moment. What is the level of trauma associated with placing a seat belt on a child? It's very minor. So if you develop bruises with that level of minor injury, it's easy to assume that something a little stronger than that could produce a major bruise. ...

So you reviewed the medical evidence in the Ernie Lopez case. Did you get a fee for that or any compensation? I mean, some people talk about experts, and they'll say, "These guys are hired guns."

Yeah. I have to say that I feel so bad, because these people are falsely accused, and they've had to pay for their legal defense. I've never charged anybody anything. One family was kind enough to give us the Marriott Miles for a family vacation.

But I asked the father, "How much did it cost you in legal expenses for this case?," and he said, "One hundred thousand dollars." I said, "Then you should pay me nothing." We do it in the hospital all the time. In my role, I'm salaried. Whether I see one patient or 100 patients, I get paid what I get paid.

I think it's just too much to think that somebody should have to have one more insult to being falsely accused, and that is to pay for this kind of support. We should do it as a public service. We the physicians need to get together in a room, talk about the case, reach a consensus. And then somebody sends you a thank-you note, and everybody feels better because the right answer was reached. ...