- NF-1 AND LEARNING DISABILITIES (continued)
Martha Bridge Denckla, M.D.
-
In our study we had nine parents with NF- 1 and no UBOs. At the time we had 20 children of
which two had no UBOs. The basal ganglia is the primary and most common site to have UBO's.
Then there are cerebullar UBOs. followed in frequency by a UBO brain stem. The main point
here is that they are all sub-cortical. Basal ganglia, cerebellum, thalamus, brain stem, all are
sub-cortical
Now, another study that we have done, but not yet published, was done by Dr. Steve Mott, a
protégé of Dr. Roger Packer. He worked with us for a year and has continued on with this
project. He had the brilliant insight to point out that the basal ganglia are connected to the
opposite cerebellum. They are very important, we know, working together as a pathway system.
So what if we made a designation of left basal ganglia to its opposite side and called that the "left
pathway" and call the opposite the right pathway? So what you need to see here is, Mott is
calling this the right pathway, but it includes the right basal ganglia and its connected left
cerebellum.
In classical literature the JLO is considered to be a "right side of the brain" task. In the JLO
you are judging the angular displacement of these lines. We are looking for a difference score.
We are going pair by pair by pair by pair -- a person with NF...has a score and the sibling has a
score. You subtract the two and get a difference score. "Lowering" JLO, as with lowering IQ.
Some people have more of UBOs on the right than the left, If they had more on the right, we
called the "right preponderance pathway group." If you look at whether they were left UBO
volume predicted by the JLO difference score, it was not.
It is very right-left confusing, but the punch line is that the right pathway UBO volume, the
total correlates with the lowering of JLO. So it acts as though it were the right cerebral cortex
even though it is not. The right pathway goes with the test that would be expected if that
pathway from the right basal ganglia to the left cerebellum acts like a pathway that is connected to
the cerebral cortex on that right side.
That is kind of a tricky concept. But again, the whole idea is, even if you have preponderantly
less UBOs, it is the right sided or right-connected-pathway ones that correlate with the difference
from your sibling. In this case we do see a volume effect, but only if we are smart enough to have
a hypothesis about what tissue is involved in a very particular task as opposed to IQ which is
supposed to be multi factorial and "diffuse" in its brain implications.
These are sub-cortical structures that people do not really think about when they refer to
things like spatial ability or any kind of intellectual cognitive ability, which makes the work very
interesting. We are emphasizing that some of these UBOs underneath the cortex in parts of the
brain usually thought of only in terms of motor control are players in cognitive tasks.
When I went to medical school these were motor parts of the brain. Nobody talked about
anything like spatial ability with these parts of the brain, basal ganglia and cerebellum which were
for movement but nothing to do with cognition. Our research on NF-1 is now contributing to
people realizing that these "motor" parts of the brain are players in cognitive activities.
The interesting thing is that we also knocked the theory a little bit on its back, because
everybody thought JLO is spatial, and spatial should predict math. Neither in the kids with NF-1
nor in their siblings is there a significant correlation between JLO and the calculations.
Not everybody who volunteered got into the study. We were very, very selective. We do not
have people with optic tract involvement because we just did not know what to expect about
perception with the optic tract involvement.
We do not take anyone who has seizures, because effects seizures medication start to interfere
with your interpretation of your results also. Fifty percent of those who talked to us about getting
into the study were receiving some special education. Sixty-seven percent had evidence that they
had reading and written language, but fewer had math problems. In other words, this is a much
bigger percentage of verbal or language-based academic problems seen in children with NF-1.
Let us remember there are many language and verbal tests and only two visual-spatial -- the
block design and JLO impaired. And spared relative to siblings most are what we call "executive
functions" which has to do with planning, organizing, being able to self-monitor. A lot of very
important stuff was not impaired. There is excellent visual discrimination and object perception.
So a lot of non-verbal functions were all fine.
What we are doing now in our second period of funding, which the National Institute of
Neurological Disorder and Strokes is providing, focuses on longitudinal changes over time in
children and adolescents with NF- 1. (The National Institute of Child Health and Human
Development is still finding several other LD centers, by the way.)
We still have these several populations that we are studying (Fragile X) (Tourette Syndrome,
ADHD) which gives us the ability to cross-compare these two neurogenetic projects. All of these
syndromes are said in the literature to have non-verbal learning disabilities. We have got some
pretty solid evidence that Fragile X does, but NF- 1, as you see, is not panning out. The others
have more executive dysfunction but NF-1 does not.
So we have a direct comparison that we can make across these neurological conditions. We
have controversial issues to deal with still: Is NF a model for developmental delay? As I said to
you before, if there are things (UBOs) inside the head that go away and people get smarter, that
would be the model for developmental delay. Something inside the brain is getting better, and
people are getting smarter and more effective. So we are really excited about being able to
demonstrate that. The question remains, does it disconfirm Rourke's model for academic and
social outcomes? We think it disconfirms his model. The reason is that with NF-1 the JLO is
bad, the Block Design is bad, so undoubtedly there is spatial disability but the NF-1 participants
do not have the outcomes that Dr. Rourke predicts they should have, such as math and all these
emotional problems. And the Fragile X group, which is much more verbal is turning out to have
the projected developmental path.
So, oddly enough, even though they both have the same problems with a different profile of
assets, they come out with a different ultimate result.
The rationale for what we are doing now is, we are trying to have people come in
longitudinally (over a long period of time). Some of you may be among those we have asked to
come in briefly but once a year to look at some of the impaired functions and some of the spared
functions for this sample.
We are doing some fancy new kinds of things that are called "longitudinal data" now since we
want to know what happens to the IQ, what happens to the UBOs, what happens to some of the
JLO and other deficits. How do these three things relate to each other over time? In other
words, does the cognitive go uphill while the UBOs go downhill? What is the ultimate fate of
UBOs? Also, do our findings hold up? Can we replicate it again in a new sample? So we do
need a new sample.
We are also looking for young adults. I'm putting out a plea here. We are looking for anyone
between the ages of 28 and 39 who has NF-1 and has an unaffected sibling who is in that same
age range. We need 15 pairs. We would like to see whether there is less of a gap between them
and their siblings when they are in that young adult age group.
We would like a better diagnosis this time around of formal questionnaires for ADHD and
social development. We also want a better pursuit of who is LD and ADHD in the family who
does not have NF- 1. After all, people don't have just one gene. As big as the NF-1 is, it is not the
only one that they have.
We will continue with sibling pair design in spite of people often attacking us about it. I was
convinced by Roswell Eldridge and I have had an excellent statistician confirm that this is
absolutely the way to tell what is going on. Impairment will continue to be defined relative to the
sibling. If we are really in a crisis and we cannot get enough siblings, we control for the mid-
parental cognition. We can always take mom's test and dad's test, take the average of the two and
say that is what we would expect of the child. We can take this procedure as a fall-back position
if we do not have a sibling. (Sometimes you cannot find both parents either, so it is not a magic
answer! )
We think that the UBO number and volume are candidates for different kinds of cognitive
underpinnings for the learning disabilities. We are getting revved up to do functional magnetic
resonance imaging with a form of JLO that we have on computer which we can do inside the
magnet, inside the machine. So we are going to start doing functional testing to find out, what is
going on in the brain. Are they not activating the same places they ought to be activating when
they encounter this test that they do so poorly on?
Just as a reminder, the main thing about the statistics is that they are the matched pairs design -
- It is always the person with NF and the sibling. It is not a group thing. It is not like the mean of
this group and the mean of that group. They are actually called matched pair statistics, which we
think is the key to understanding these issues.
Any of you who are interested either for children, pairs or the young adults pairs, we would
be very, very happy to talk to you and get you involved. You can call our study coordinator,
Chris Koth, at 410-955-9399.
Q: When you refer to children, "outgrowing the LDs and the adults getting smarter''
could it be that they are just finding a different way of coping with those LDs?
A: Sure. Certainly. That is why you need an elaborate kind of statistics with a comparison.
Some of the siblings have LDs and we can see whether it is any different between the
people with NF-1 and the siblings who just happen to have some LDs quite apart from any
NF issue.. So we have a built-in comparison.
Q: At what age do the UBOs start disappearing?
A: Late. We took an arbitrary cut point of age 15 when we did the study and surveyed the
hospital records. It seemed as if we could not find anyone who was 16 or over 16 who
had UBOs. Now the late Wes Fox told me once in a meeting like this that he was pretty
sure he had UBOs. But I never got a chance to really look into that. He might have been
thinking retrospectively. But we have a strong consensus from people we talk to around
the country from the Texas NF clinic at M.D. Anderson ... that they do not see UBOs in
the late adolescent and certainly not in adults.
Q: What is the oldest age of children that you have been testing? So they have until
now, been under 18?
A: Yes. Some of our original sample are sneaking up into that age group now. But they
would not be included this new sample. They would be in our longitudinal follow-ups.
Originally, I think the oldest person we took was 14 at the very beginning. We have
different questions to ask in the longitudinal portion of the study.
Q: On the longitudinal study are you trying to see whether the IQs had come up and if
the methods of coping have overcome these kinds of questions?
A: We have ways of getting at that too. We use such extensive language in neuropsych
testing that we can determine whether there are compensatory factors or whether the raw,
natural power is better in those areas. We can tell the difference.
Q: I knew when my daughter was in second grade that she had learning disabilities.
But she did not fit into any kinds of disabilities that they could test for back in late
'70s, early '80s. All things that you are telling me were the problems that she had,
but she did not fit any of these profiles.
A: Shall I tell you the biggest reason that the children don't fit? It is a very common reason
that I see week in and week out in my own clinical work. The problems that they have
impact both the IQ and the academics. The schools, in Maryland anyway, use a criterion
which is a discrepancy. It is illogical biologically to ask for the discrepancy if your
underlying problems are pushing down both variables. How are you going to get a
discrepancy? In other words, if you have some brain based language problems, and they
are going to lower your verbal IQ and your reading the way I showed you, how are you
going to be defined as learning disabled? They are just going to say, "you are not learning
disabled. You are reading as well as can be expected for your IQ." And that is the kind of
run-around that is not unique to this group. It happens to lots of children, and I tear my
hair out trying to explain it to school systems all the time.
Q: The other thing is though, they told me that her IQ was fine, but she was
compensating on her own so it would be no good working with her.
A: Parenthetically, nobody in this whole study had a below-normal IQ. Everybody had a
normal IQ. So we are talking about just being lower than your brother or sister. Your
child may be compensating on her own, but is that too exhausting to keep doing?
Q: You ruled out people with optic gliomas.
A: Yes. People have challenged us on that. It is only to be purists and to be sure we do not
get into arguments that anything visual possibly could be a hindrance, because the biggest
disability has to do with visual-spatial. People would ask us, "how do you know?" It
could be visual acuity's fine, but how do you know it does not impact in some subtle way
on the brains ability to perceive things? I think I'd be a lot braver in a few years to say,
that I will take anybody as long as the optic glioma is stable and it is not interfering with
vision.
page 1 | 2
|