We can take the concept of levodopa equivalent dose (LED) a step further by taking into account the length of time over which a dose is effective.

The length of the effective period depends in part on a drug's half-life, but it also varies from person to person. For instance, depending on the number of dopamineric vesicles remaining, and the intensity of exercise. My estimates of the length of the effective period for me are:
rasagiline (Azilect), 24 hours,
ropinirole extended release (Requip XL), 24 hours;
Stalevo, 3 hours
Sinemet, 3 hours

A rough measure of the potency of a Parkinson's drug is given by its LED/hr:
1mg rasagiline has a potency of 100/24 = 4mg LED/hr
1mg ropinirole extended release = 16.67/24 = 0.7mg LED/hr
100mg Stalevo = 125/3 = 42mg LED/hr
100mg Sinemet = 100/3 = 33mg LED/hr

These figures show, for instance, why many people don't notice the effect of rasagiline: its hourly LED is less than one eighth of that of Sinemet. It does have the advantage of keeping LED/hr levels consistently higher during the whole 24 hour period, thus reducing the impact of an "off" and reducing the chance of levodopa induced dyskinesia.

The next stage is to graph the changes to the hourly LED during the day taking into account the time it takes for a pill's levodopa to reach the brain. This is affected by gastric emptying and competing protein transport through the blood brain barrier.

I've no empirical data for this, but one would expect a graph like this for a person taking a mixture of medications.


LED/hr
|-------------------------- levodopa induced dyskinesia
|.....................
|.....................
|.....................
|.....................
|.....................
|.....................
|..........SSSSSSSSS
|..........SSSSSSSSSS
|..........SSSSSSSSSSS
|.........SSSSSSSSSSSSS
|.........SSSSSSSSSSSSS
|.........SSSSSSSSSSSSS
|-------------------------- on-time threshold
|.........SSSSSSSSSSSSSS
|........SSSSSSSSSSSSSSS.
|RRRRRRRRRRRRRRRRRRRRRRRR
|RRRRRRRRRRRRRRRRRRRRRRRR
|AAAAAAAAAAAAAAAAAAAAAAAA
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
|EEEEEEEEEEEEEEEEEEEEEEEE
--------------------------- Time(hr)
|000000000011111111112222
|012345678901234567890123

Example: LED/hr graph
E=endogenous dopamine production, a=Azilect, r=ropinirole, s=Stalevo

The graph makes it clear why moving from Sinemet to Stalevo could lead to levodopa induced dyskinesia: the 25% extra strength could take one over the LID threshold.

Analysing such graphs offers a way of estimating the rate of production of endogenous dopamine (dopamine produced naturally) that still remains. If this is measured over time, the changes can be used to give an estimate of disease progression.

John