Q: If Lithium Works, You Have Bipolar?
I have a friend who has been diagnosed with bipolar II -
she was told that the litmus test for this diagnosis is that lithiam
straigthened her out during a manic episode - she was told that if the drug
works, you are definiely bipolar - she suspects that she is not bipolar because
her hypomania episodes were always drug induced. She desparately needs
treatment for depression, but her dr. won't give her antid's because of the
bipolar - but if she is not bipolar she is being seriously short-changed. It is
my impression that there is no "litmus test"for bipolar - can you comment?
Hello Barbara --
In my opinion... I think
it's safe to say that "arriving at a diagnosis based on response to a
medication" is not accepted as a diagnostic method. One hears words to
the contrary from diagnostic logicians quite frequently. That much is pretty
Moreover, in this particular case, we know that lithium
can be effective in unipolar depression. So in this particular case, her
response clearly does not clinch "bipolar". I'd say that position also
would be widely accepted.
Now for the "but"... let's think about how "seriously
short-changed" your friend might be depending on how one views her diagnosis.
First, imagine she's "really" (by some magic diagnostic means of knowing, as if
there were such a thing, which is not at all clear until we have better biologic
markers for distinctions like "unipolar" versus "bipolar") unipolar, not bipolar
(one more caveat: mood experts generally agree this is a continuum; not a
black/white distinction, as though one could be one, and "not" the other, which
is misleading in the very language).
There, if I haven't confused things yet, let's proceed.
Imagine: by the magic means, your friend is "unipolar". We make a mistake: we
call her "bipolar". She gets treated with lithium. She gets better anyway, as
you've seen. Now, is there any evidence that making a mistake in this way could
harm her? I.e. that lithium could make unipolar disorder worse? No. Side
effects, long term risks, yes; but fundamentally worsening the condition, no,
nor is this something where you find mood experts writing about such
By contrast, flip the scenario the other way: she's
really "bipolar". We make a mistake: we call her "unipolar". She gets treated
with an antidepressant. In this case, she's already gotten worse. Is there any
evidence that making a mistake in this way could harm her more permanently? Not
much, granted, but just a bit (e.g. the Stanford study by
and Ketter). And this is an issue you see mood experts writing about all
So, before we conclude she's being potentially
"short-changed" by the caution about re-introducing antidepressants, I offer the
above analysis as suggesting that she may be potentially protected by such a
view, at least as much as short-changed.
To illustrate this principle just one more way, check
this out: that same group at Stanford just published another study using this
same logic: they studied adolescents who had mild symptoms of
depression, not manic symptoms -- but who had at least one bipolar
parent. They gave these kids Depakote for their depressive symptoms, instead of
an antidepressant. Depakote is a mood stabilizer like lithium, as you know,
but one that is not generally regarded as having inherent antidepressant
effects. As you can see in the
for this study, 78% of those who completed the 3-month study (one dropped
out at 2 weeks) were improved or much improved.
I think that study has some serious implications for
people who have depression symptoms and some reason to suspect some degree of
"bipolarness" as well. My opinion. Good that you're asking such questions,
though; it's good for everyone to have to think out loud about this kind of
Published March, 2004