Aug 18 2008

Schiavo and the Persistent Vegetative State

Published by under Uncategorized
Comments: 15

This is my second entry discussing the issue of persistent vegetative state and Terri Schiavo. Actually it’s the third – the first was a review of a newly published study of the poor news coverage of the Schiavo case. In response Dr. Egnor wrote a blog entry (although he didn’t actually respond to any points I made in the post, it looks like he was just itching for an exchange on this issue), offering to discuss the relevant issues in our respective blogs. I wrote the first part of my response Friday, and here is part II.

In this entry I will review the medical facts of the Schiavo case, as best as I can reconstruct them. I was never directly involved with the case, I never examined her or reviewed original medical documents (except those made public, like the autopsy report). This is a minor problem, of course, as I must depend upon the examination of other neurologists. So to be clear I am not offering a direct medical opinion in this case – I cannot do that never having examined her myself – but rather an analysis of the public documents in the case.

To quickly review the medical history, Terri Schaivo collapsed at her home on February 25, 1990. Her husband, Michael, was home with her and immediately called EMS. Terri had a respiratory and cardiac arrest, although the exact cause of that arrest was never definitively determined. She was revived but suffered a diffuse anoxic injury (her brain had insufficient oxygen for a prolonged period of time) leading to extensive damage to her brain. For the next 15 years she was in a comatose state – a state of decreased consciousness and neurological function. Her case came to national attention over the controversy of her treatment. Her husband claimed that she had expressed the wish not to be kept alive in such a state. Her parents insisted that her religious views were such that she would want to be kept alive. Eventually her husband prevailed. Feeding and hydration were withdrawn and she passed away on March 31, 2005.

The political and ethical controversy centered around the right to life vs the right to die, and the proper roll of the various institutions of government vs the right to privacy in making such decisions – with high passions on all sides. My purpose here, however, is to focus on the medical controversy underlying the case. As I wrote previously, the scientific/medical questions of the case were distinct from the political and ethical questions. In an ideal world, the science would be determined objectively and dispassionately so that it can properly inform the ethical debate. While that may ultimately have occurred, at least in a subset of the information presented at court, much of the courtroom testimony and most of the public debate was marked by the medical facts being twisted to ideology. Even today it is extremely difficult to find sources on the Schiavo case that are not heavily biased.

I will try to break down the relevant facts of her medical condition near the end of her life.

Neurological Exam

This was the most contentious piece of information in this case. The question raised was whether or not Terri Schiavo was truly in a persistent vegetative state or was she rather in a minimally conscious state. PVS allows for no evidence of conscious activity, although automatic reflexes remain. A minimally conscious state allows for minimal evidence of consciousness. This diagnosis was not created until after Schiavo was initially diagnosed with PVS, although it was well established by the time her controversy became public.

The difference, as the names suggest, is subtle – no consciousness vs minimal consciousness. Both result from significant neurological damage. There is a slight difference in prognosis (although other factors are also very important, such as duration – the longer some one is in either state without improvement the worse the prognosis). Patients in PVS almost never recover, and arguably the few rare cases may have been misdiagnosed cases of minimally conscious state. Whereas those in a minimally conscious state have a very small chance of improvement, which can be significant, although once chronic they never return to normal neurological function.

This subtle, but definitively meaningful, distinction was what the medical controversy focussed on.

The case attracted a number of physicians to offer their expert testimony for the various court trials deciding Schiavo’s fate, and to offer their unofficial opinions for either side of the case. It appears that many of the physicians who offered their opinions fell into one of two categories – they had a strong ideological bias going in and their medical opinions matched their bias, or they were charlatans and self-promoters exploiting the case for a little name recognition.

I already described my opinion of William Hammesfahr, who was the primary neurologist for the parents. He was promoting his own vasodilator therapy – a quack therapy with no evidence of efficacy. He offered a number of dubious opinions in this case, but most importantly that his therapy could help Terri Schiavo. The other physician who gave official testimony for the parents’ side (That Schiavo could recover) was Dr. William Maxfield. Like Hammesfahr, he also has a dubious treatment to sell – hyperbaric oxygen. Also of note Dr. Maxfield is a radiologist – not a neurologist. I find it extremely telling that the two physicians who ultimately stood for the position that Terri was not in a PVS had something dubious to sell.

There was another physician of note, a neurologist and recognized expert, who publicly offered his opinion in the case – Dr. William P. Cheshire. He stated that he believed Schiavo exhibited signs of consciousness. His opinions were largely criticized, however, in that they were based largely on his “feelings” about Schiavo, rather than objective exam findings, and he also came to the case on record as having strong conservative religious views on the topic.

A number of neurologists examined Terri Schiavo over the years and they all felt, to a high degree of confidence, that she was in a PVS. Most notably was Dr. Ronald Cranford, a noted expert on PVS. He took a great deal of heat from the conservative media for his medical opinions, and in some interviews became noticeably peeved. His report of his exam, however, sounds very credible. I am also a clinical neurologist, so I have a sense if someone understand the key issues of such an exam, and Dr. Cranford displayed a firm grasp of the neurological issues (in fact he is more of an expert than I).

The primary concern, in terms of making the clinical diagnosis of PVS, was whether or not Terri Schiavo displayed any clear signs of conscious awareness or interaction with her environment. There were many reports of her smiling or laughing in response to what was being said around her, protesting unpleasant procedures, and looking people in the face. However, it is very common, because it is extremely easy, to misinterpret the random actions of someone in PVS as being purposeful. The situation is a setup for confirmation bias. For example, if a patient in a PVS lets out a guttural sound an observer may conclude that she must be upset at something – and then find something to explain her reaction. They could then report that the patient responded in protest to whatever it was that they retro-fitted to their action.

To control for such “overcalling” of responsiveness, a patient must be observed for a suitable period of time to see if they consistently respond to any stimuli. The response has to be reproducible (not absolutely consistent, just reproducible enough to say that it is a true pattern and not a coincidence). The consensus of opinion of neurologists who examined Terri Schiavo is that she displayed no consistent pattern of responsiveness – which is consistent with a PVS.

The other side of the coin, of course, is the question of how to avoid undercalling conscious signs. The answer is that the standard neurological exam includes using multiple methods to attempt to provoke a response: visual, auditory, tactile, and even painful, combined with prolonged observation of spontaneous behavior, and a thorough understanding of brain reflexes.

The next obvious question is – how sensitive is this neurological exam for the detection of subtle consciousness. The problem with this question is that we do not have a gold standard to compare the exam to. If we did then we would rely upon that gold standard, whatever it was. What we can say is that the clinical diagnosis of PVS, based upon the thorough standard exam, is a reliable prognostic indicator. If someone meets criteria for a PVS then they have a prognosis for recover of almost zero. As with all of science, we don’t get to peek at the absolute answer. There is no teacher’s edition to the universe (I forget from whom I am stealing that quote). All we can say is how well our models make predictions. The PVS diagnosis is a very good predictor of outcome.

Internal consistency is another criterion to which we can subject our scientific models. In the case of PVS we can ask – how well does the clinical exam match measures of brain damage or other measures of brain function.

EEG and CT scan

Terri Schiavo had a CT scan of the brain in 1996 and in 2002. Here is the scan from 2002. The black spaces are cerebrospinal fluid. The gray areas are brain tissue. The picture shows a normal brain on the left and Terri Schiavo’s brain on the right. The scan clearly shows that she had lost a profound amount of her brain tissue, which had been replaced by fluid (a finding called hydrocephalus ex vacuo).

The CT scan shows anatomy. We can also use EEG (electroencephalogram) as a gross way of looking at the electrical activity of the brain – function. Dr. Cranford testified that Terri Schiavo’s EEG showed no cortical activity.

Therefore, the exam, the CT scan, and the EEG all roughly agreed with each other – they each were independent evidence for profound diffuse cortical injury consistent with a PVS.

Despite this evidence the medical controversy continued. The arguments on the other side were not very compelling, however. Some noted that the CT scan, while profoundly abnormal, showed some improvement from 1996 to 2002. I could not find a direct comparison. However, given the severity of the 2002 scan any possible amount of improvement would not alter the ultimate final diagnosis or prognosis.

Others criticized the fact that Terri Schiavo did not have an MRI scan, which is more detailed. However, an MRI scan could not change the facts obvious from the CT scan – that there was profound atrophy. The MRI scan could not have shown less damage – but it could have shown more damage. It may have shown that some of the gray matter visible on the CT scan was actually scar tissue, not living brain tissue. It could only have worsened the diagnosis, not improved it.

Dr. Cranford also noted that an MRI scan was not possible in this case because Schiavo had intrathalamic stimulators – electrodes in her brain. This metal could not go into the powerful magnet of an MRI scan. This was actually the first I read that Schiavo had intrathalamic stimulators in place. The only references I could find to this are ones mentioning that it was a contraindication to MRI. I will have to look further into this. The only indication I can think of for Schiavo to have had these stimulators was as an experimental treatment of her PVS. But I could not find any reference to answer this question. If someone knows of one, please pass it along.

Still others called for a PET scan, which measures blood flow to the brain. This would not have been a useful test, however. Even if it showed that all of the brain tissue seen on CT scan was alive and active brain tissue, a best case scenario, that would not have altered the diagnosis or prognosis. Like the MRI scan, this information could only have worsened her diagnosis (showing some brain tissue was not active), not improved it.

The Autopsy

The official autopsy report indicates that Schiavo’s brain weighed 615 grams. This is about half the weight of a healthy woman of her age and size. This very straightfoward and objective finding confirmed the CT scan and is consistent with the diagnosis of PVS.

One question that emerged after the autopsy results were made public was whether or not Terri Schiavo’s extreme state of dehydration (the proximate cause of her death) reduced the size of her brain at autopsy. This is a reasonable question, and fortunately we have specific information to answer it. The brain itself experiences only minimal loss of water during dehydration. Most of the tissue fluid lost comes from the skin and muscles. Even if we assume a greater fluid loss than average, dehydration could not come close to explaining this degree of volume and weight loss of her brain.

Also – the pathologists did more than just weigh her brain. Much of the media reporting of the autopsy focused on the weight and the dehydration question, but this is rendered moot by the pathological findings. If you can parse the jargon, read the autopsy report I linked to above. The pathologist, Dr. Thogmartin, wrote about her brain, “The changes seen were striking in their appearance and global in their distribution.” Here is my summary:

The autopsy showed that Terri Schiavo had diffuse watershed infarcts of her brain – this means that the borderlands between adjacent arteries, where perfusion pressure is the lowest, experienced the most profound loss of oxygen and therefore damage. This is consistent with her history of cardiac arrest and diffuse anoxic-ischemic injury. The cortex was everywhere very thin, but worse in the occipital lobes. Microscopically her brain showed great variability. The worst areas showed a near complete loss of neurons – the cells primarily responsible for brain function. The better parts showed “relative” preservation, but still had loss of some types of neurons and infarction of some layers of the cortex, as well as overall atrophy.

In other words – her brain showed severe chronic damage, grossly and microscopically. The brain seen on autopsy was probably incompatible with any detectable consciousness and would have had no hope of meaningful recovery. Any other opinion is pure pseudoscience.

Of note the damage was worst in the occipital lobes responsible for vision. It is therefore also clear that Terri Schiavo was blind. This is further and independent evidence that any reports of her fixing her gaze on people or recognizing her family by sight must have been false.


Terri Schiavo’s case is certainly a sad and tragic one. While the ethical and legal issues of her case remain controversial, there is no genuine scientific controversy about the medical facts of her case. The autopsy put an end to any legitimate dissenting opinion regarding her neurological condition.

15 responses so far