Back to the Top
My quick survey of literature seems to suggest that for non-steady-state PK/PD modeling, arterial PK sampling is always preferred vs. venous sampling. Can someone shed some light on if this is the case if the PD effector site is in peripheral tissues? Would drug concentrations in venous blood tell a little bit more accurately the "real time" tissue concentrations? It is preferred if someone knowledgeable can provide a conceptual discussion (rather than a list of publications) on the topic, without taking regard of operational easiness to collect blood in venous veins.
Regards,
Jack
Back to the Top
The following message was posted to: PharmPK
Hi Jack,
I've done a number of studies with arterial and venous blood samples and PD measurements. Arterial blood is the blood supplying most organs of the body. A PD effect is best related to arterial blood concentration, as any effect delay (hysteresis) will generally reflect the rate of target organ/site equilibration (assuming rapid receptor dissociation).
e.g. Effect delay = target organ delay
Of course, it is not always possible to draw arterial blood. The thing to remember about venous blood is that it is blood draining one or more organs of the body - typically the arm in clinical studies. For some drugs, there is a significant equilibration time across the arm, and venous blood concentrations will lag behind the arterial. This makes the effect delay seem smaller, but the relationship between venous blood and effect is more convoluted:
e.g. Effect delay = target organ delay - arm delay
If your venous blood sampling cannula is in the vena cava and is partially sampling drug depleted blood downstream from the liver, or drug enriched blood downstream from an infusion or other administration site, things get even more complicated.
Hope this helps - not a easy topic to summarize in a paragraph or two!
Regards,
Richard Upton
Back to the Top
The following message was posted to: PharmPK
Jack,
I guess, arterial sampling is preferred for the PK-PD modeling of the very fast-acting drugs (anesthesiology would be a good example). Indeed, equilibration time (when arterial and venous drug concentrations become close if not identical) is just 2-3 minutes. If the PK-PD delay (from the IV dose to the effect) is less than these 2-3 minutes, you may want to use arterial samples: otherwise the effect may precede concentration changes. If the delay is more than 3 minutes, I would not expect any significant advantages of the arterial sampling.
Regards,
Leonid
Leonid Gibiansky, QuantPharm LLC
www.quantpharm.com
Back to the Top
The following message was posted to: PharmPK
To follow up the messages from both of you (Richard and Leonid), is it fair to say any potential arterial-venous difference is mainly of some interest if and when one observes a proteresis (then investigated as one of the possible causes)? Also, even at PK steady-state, you can't expect arterial-venous difference to completely go away, can you?
Jack
Back to the Top
The following message was posted to: PharmPK
Dear Colleagues:
When a drug can be radiolabeled with a positron emitter, such as 11C, 13N or 18F, the wealth of information that can be gained is orders of magnitude greater that a one (two?) time sampling from a specific vein and/or a specific artery.
1. Regions-of-interest can be defined SIMULTANEOUSLY for arteries and veins at different sites, for selected tissues, etc
2. Data can be obtained for periods of the order of seconds, allowing HUNDREDS of data points
3. The same subject (animal, human) can be studied more than once, especially at one or more time points following treatment
4. It is noninvasive
Like all methods, dynamic imaging studies have limitations
a. Preparing a PET labeled molecule is not trivial
b. Human studies require an IND or an RDRC approval
However, having good PET data from a representative molecule of a family will allow more conventional studies to be performed in a much more rational manner.
Back to the Top
The following message was posted to: PharmPK
Hi Jack,
To pick up this conversation again. If we are happy with the idea that proteresis between effect and the concentration at the site of actual site of drug action is impossible, then an observed proteresis would most likely be an artefact of venous blood sampling.
Walter, I agree that imaging studies are an elegant solution to informing the physiological basis of drug disposition. My colleagues and I have used blood flow measurements combined with blood sampling from multiple sites in the body for the same purpose (in animals and man). Drug kinetics in individual organs can be inferred in vivo from sampling afferent and efferent blood. Not as elegant, but low tech and achievable with modest equipment. I think it is an under-utilized technique.
Leonid, I agree that for some drugs difference in arterial and venous blood can be ignored - e.g. drugs that distribute into extracellular space. However, in my experience (and I have mostly worked with anaesthetic related drugs) lipophilic, basic drugs can have prolonged uptake into the lungs and peripheral tissues which continues at even at pseudo-steady state (i.e. when the blood concentrations appear constant). For example, complete equilibration of fentanyl across an arm takes about 10 hours. These drugs can have large differences between arterial and venous kinetics, and it is very easy with a poorly designed study to attribute tissue uptake to clearance.
There is definitely more to learn about the relative information gained from arterial and venous samples, and the use of techniques like "arterialized" samples achieved from local heating.
Regards,
Richard Upton
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Arterial versus venous sampling" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)