ssussu
☆    

China,
2018-02-06 11:48
(2242 d 01:34 ago)

Posting: # 18362
Views: 10,182
 

 When the the infusion speed deviate from the protocol specified [Study As­sess­ment]

Dear guys,
I am writing this post to consult a problem.
We execute a drug clinical trial, the drug is specifed to administrate by infusion for 0.5 h as a same speed. We set the infusion speed as calculate, but unfortunately, during the infusion, we found a bubble in the infusion tube. For safety, we have to stop the infusion and expel the bubble in the infusion tube, and this stop takes 5min. And to complete the infusion in 0.5h, we speed up the infusion pump speed.

We will assessment the PK for this drug by NCA. What's the effect of this situation? Does the Cmax should be excluded? What about the AUC? What should the investigator to do if the same situation (need to stop temporarily or need to change the infusion speed) occurs next time?

Thank you.


Edit: Category changed; see also this post #1. [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-02-06 14:28
(2241 d 22:54 ago)

@ ssussu
Posting: # 18364
Views: 9,389
 

 Different PK profile

Hi ssussu,

❝ […] during the infusion, we found a bubble in the infusion tube. For safety, we have to stop the infusion and expel the bubble in the infusion tube, and this stop takes 5min. And to complete the infusion in 0.5h, we speed up the infusion pump speed.


Be aware that the dose will not be correct since likely you lost a part by flushing the tube. Whether this is really relevant depends on the lost volume / total infusion volume.

❝ We will assessment the PK for this drug by NCA. What's the effect of this situation? Does the Cmax should be excluded? What about the AUC? What should the investigator to do if the same situation (need to stop temporarily or need to change the infusion speed) occurs next time?


I would stop the infusion and exclude the subject. Reasons:
  1. During stop, elimination continues, and
  2. only if you lost nothing (almost impossible) the Cmax and AUC will equal the one with a constant rate infusion.
I don’t see a solution if you will analyze the data by NCA. If you are able to measure the lost volume, you could analyze the data by a PK model correcting the doses and adjusting the infusion rates.
Example: t½ 8 h, infusion interrupted after 15 min for 5 minutes, adjusted infusion rate, no dose lost.

[image]


Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ssussu
☆    

China,
2018-02-07 03:50
(2241 d 09:32 ago)

@ Helmut
Posting: # 18365
Views: 9,348
 

 Different PK profile

Hi Helmut,

Thanks for your reply.

❝ Be aware that the dose will not be correct since likely you lost a part by flushing the tube. Whether this is really relevant depends on the lost volume / total infusion volume.


The dose is not lost because the nurse expel the burble by squeezing the tube to float the burble up from the liquid surface of the tube.

❝ 2. only if you lost nothing […] the Cmax and AUC will equal the one with a constant rate infusion.


❝ Example: t½ 8 h, infusion interrupted after 15 min for 5 minutes, adjusted infusion rate, no dose lost.


[image]



From the above picture, if the dose is not lost, can i say just the AUC will be effect but the Cmax is the same with the canstant rate infusion?


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-02-07 12:04
(2241 d 01:17 ago)

@ ssussu
Posting: # 18368
Views: 9,292
 

 Different PK profile

Hi ssussu,

❝ The dose is not lost because the nurse expel the burble by squeezing the tube to float the burble up from the liquid surface of the tube.


I don’t understand. Did you use an infusion pump or an infusion bottle / drip chamber?

❝ ❝ 2. only if you lost nothing […] the Cmax and AUC will equal the one with a constant rate infusion.

❝ ❝

❝ ❝

[image]



❝ From the above picture, if the dose is not lost, can i say just the AUC will be effect but the Cmax is the same with the canstant rate infusion?


Read again what I wrote above.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ssussu
☆    

China,
2018-02-07 12:42
(2241 d 00:39 ago)

@ Helmut
Posting: # 18369
Views: 9,265
 

 Different PK profile

Hi Helmut,

❝ ❝ The dose is not lost because the nurse expel the burble by squeezing the tube to float the burble up from the liquid surface of the tube.


❝ I don’t understand. Did you use an infusion pump or an infusion bottle / drip chamber?


Yes, the burble is extruded by squeezing the tube to float the burble up from the liquid surface of the tube, like this
[image]
so, the dose is not lost.

❝ ❝ ❝ 2. only if you lost nothing […] the Cmax and AUC will equal the one with a constant rate infusion.

❝ ❝

❝ ❝ […] if the dose is not lost, can i say just the AUC will be effect but the Cmax is the same with the canstant rate infusion?


❝ Read again what I wrote above.


So, you mean the Cmax and the AUC both will equal the one with a constant rate infusion? Don't we need to consider the eliminate rate? if we speed up the infusion rate while the eliminate rate is not change, the Cmax won't be greater than the constant rate infusion?


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-02-07 15:13
(2240 d 22:08 ago)

@ ssussu
Posting: # 18371
Views: 9,424
 

 Suitable equipment?

Hi ssussu,

❝ ❝ […] Did you use an infusion pump or an infusion bottle / drip chamber?


❝ Yes, the burble is extruded by squeezing the tube to float the burble up from the liquid surface of the tube, like this

❝ [image]


❝ so, the dose is not lost.


❝ ❝ ❝ ❝ 2. only if you lost nothing […] the Cmax and AUC will equal the one with a constant rate infusion.

❝ ❝ ❝

❝ ❝ ❝ […] if the dose is not lost, can i say just the AUC will be effect but the Cmax is the same with the canstant rate infusion?

❝ ❝

❝ ❝ Read again what I wrote above.


❝ So, you mean the Cmax and the AUC both will equal the one with a constant rate infusion?


Yes!

❝ Don't we need to consider the eliminate rate?


Why? Elimination is independent from any kind of input.

❝ if we speed up the infusion rate while the eliminate rate is not change, the Cmax won't be greater than the constant rate infusion?


Again: yes.

In my simulations I assumed:
  1. ▬▬ D=1 with a constant infusion rate of 2 h–1 (i.e., infusion time of exactly 30 minutes).
  2. ▬▬ D=0.5 with a constant infusion rate of 2 h–1 (i.e., planned infusion time of exactly 30 minutes).
    Infusion completely stopped after 15 minutes for five minutes.
    Infusion of the remaining D=0.5 resumed at 20 minutes with an accelerated infusion rate of 3 h–1 (i.e., infusion time of exactly ten minutes).
I have some doubts whether with your equipment (drip counter) you managed
  1. to have the same infusion rate across all subjects (I guess in some infusion was completed earlier or later than at 30 minutes) and
  2. to adjust an accelerated infusion rate properly.
ElMaestro had valid points as well. With an accelerated infusion rate you may run into safety problems.

Coming back to your very first question:

❝ Does the Cmax should be excluded? What about the AUC? What should the investigator to do if the same situation (need to stop temporarily or need to change the infusion speed) occurs next time?


For the next time I strongly recommend infusion pumps (yeah, I know, expensive): Exact infusion rates, no bubbles, no problems.
If you would have kept the original infusion rate (after the five minutes stop), i.e., complete it at 35 minutes, you would have seen a Cmax which is 0.35% (!) lower than expected with the planned schedule – if you have an early sampling time point. If not, you will observed a lower Cmax. How much lower depends on the sampling schedule and the elimination rate. Theoretically the AUC is not affected (depends only on the dose).

[image]



PS: Why did it take the nurse five minutes to expel the bubble? In my experience about one minute should be sufficient. Would look like this:

[image]


Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ssussu
☆    

China,
2018-02-08 05:07
(2240 d 08:14 ago)

@ Helmut
Posting: # 18377
Views: 9,227
 

 Suitable equipment?

Hi Helmut,

Thank you for your patience.

❝ ❝ So, you mean the Cmax and the AUC both will equal the one with a constant rate infusion?


❝ Yes!


I thought the AUC would be effect because I saw the the black area in the below picture. Now i know the AUC is only depend on the dose as you said.
[image]

❝ For the next time I strongly recommend infusion pumps (yeah, I know, expensive): Exact infusion rates, no bubbles, no problems.


Yes, they used the infusion pumps, and I don't know why it occurs bubbles and I need to do more detailed investigation.

❝ If you would have kept the original infusion rate (after the five minutes stop), i.e., complete it at 35 minutes, you would have seen a Cmax which is 0.35% (!) lower than expected with the planned schedule – if you have an early sampling time point. If not, you will observed a lower Cmax. How much lower depends on the sampling schedule and the elimination rate. Theoretically the AUC is not affected (depends only on the dose).


Could I say (theoretically) that no matter how long will I stop the infusion,only if I complete infusion at 0.5h,the Cmax will be same with constant rate infusion for 0.5h?

❝ Why did it take the nurse five minutes to expel the bubble? In my experience about one minute should be sufficient.


Yes, actually they take 2min and I exaggerate it to 5min, want to see the effect if the stop time is not short enough to be ignore.


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-02-08 11:28
(2240 d 01:54 ago)

@ ssussu
Posting: # 18379
Views: 9,323
 

 AUC=f·D/CL

Hi sussu,

❝ I thought the AUC would be effect because I saw the the black area in the below picture. Now i know the AUC is only depend on the dose as you said.

❝ [image]


Exactly. The input function does not appear in the basic equation of PK (AUC=ƒ·D/CL). Since after an i.v. dose ƒ=1 (by definition) and the entire dose was administered, the AUC will be exactly the same regardless the input scheme.
Say you stopped after 15 minutes and administered the remaing dose after 29 minutes as a fast (one minute) bolus, it will look like this:

[image]


Note that the red line after tmax is above the blue one. With a t½ of 8 h the “missing” part before tmax will be outweighed in the elimination phase (speaking of AUC0–∞; the AUC0–2 will be lower, of course).

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ssussu
☆    

China,
2018-02-08 12:21
(2240 d 01:01 ago)

@ Helmut
Posting: # 18380
Views: 9,439
 

 AUC=f·D/CL

Hi Helmut,

As I simulate 3 condition of infusion by WinNonlin PK Modeling,
1. infusion by constant rate for 30min and complete the infusion at 30min;
2. infusion 15min and stop for 5min and speed up the infusion rate to complete at 30min;
3. infusion 15min and stop for 5min and keep the same infusion rate and complete at 35min.
below is what i get:
[image]

It's different from what you have present. Now I am confused, I don't know why.:confused:
Hope to discuss with you more.
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2018-02-08 22:55
(2239 d 14:26 ago)

@ ssussu
Posting: # 18383
Views: 9,128
 

 AUC=f·D/CL

Hi ssussu,

❝ As I simulate 3 condition of infusion by WinNonlin PK Modeling,

❝ 1. infusion by constant rate for 30min and complete the infusion at 30min;

❝ 2. infusion 15min and stop for 5min and speed up the infusion rate to complete at 30min;

❝ 3. infusion 15min and stop for 5min and keep the same infusion rate and complete at 35min.

❝ below is what i get:

❝ [image]


❝ It's different from what you have present. Now I am confused, I don't know why.:confused:


Of course it looks different. You have an enlarged time scale (one hour instead of my two) and a much faster elimination. My t½ was eight hours (k 0.0866434). Sorry to say but you should try to understand the basic concepts of PK before using such a powerful tool like Phoenix/WinNonlin. :no:

Last hint: Run NCA (lin-up/log-down trapezoidal) on the predicted data, extrapolate to infinity, and compare AUCinf of the three scenarios. What do you get?

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ssussu
☆    

China,
2018-02-09 01:31
(2239 d 11:50 ago)

@ Helmut
Posting: # 18384
Views: 9,175
 

 AUC=f·D/CL

Hi helmut,


❝ ❝ As I simulate 3 condition of infusion by WinNonlin PK Modeling,

❝ ❝ 1. infusion by constant rate for 30min and complete the infusion at 30min;

❝ ❝ 2. infusion 15min and stop for 5min and speed up the infusion rate to complete at 30min;

❝ ❝ 3. infusion 15min and stop for 5min and keep the same infusion rate and complete at 35min.

❝ ❝ below is what i get:

❝ ❝ [image]

❝ ❝

❝ ❝ It's different from (what you have present)your result about Cmax. Now I am confused, I don't know why.:confused:


❝ Of course it looks different. You have an enlarged time scale (one hour instead of my two) and a much faster elimination. My t½ was eight hours (k 0.0866434). Sorry to say but you should try to understand the basic concepts of PK before using such a powerful tool like Phoenix/WinNonlin. :no:

What I said it's different meanss the Cmax is different in the 3 condition, but not the same as your result.

❝ Last hint: Run NCA (lin-up/log-down trapezoidal) on the predicted data, extrapolate to infinity, and compare AUCinf of the three scenarios. What do you get?

Yes, AUC is the same,I mean Cmax is different.
ElMaestro
★★★

Denmark,
2018-02-07 14:20
(2240 d 23:01 ago)

@ ssussu
Posting: # 18370
Views: 9,398
 

 When the the infusion speed deviate from the protocol specified

Hi ssussu,

this is a very simple problem, and yet with no clear answer.
I find it highly interesting.

I think Helmut has some very valid good points still. Infusion stopping implies a perturbation on your data.
I would like to add one thought:
ICH E6, §2.13 "Systems with procedures that assure the quality of every aspect of the trial should be implemented (...and added recently...) Aspects of the trial that are essential to ensure human subject protection and reliability of trial results should be the focus of such systems."
My interpretation is that the Investigator (or her/his delegate) should be able to tell if this perturbation affects reliability / renders any data extracted from the subject unusable. If it does, then the ethical action is to discontinue the subject (e.g. §2.3, look also at protocol deviations). A soft scenario is to keep the subject but only using the data for tolerability, so if there was a PK, PD or efficacy component then exclude the subject from the dataset. At any rate (no pun intended), whether to include the subject or not requires sound healthy judgement and a signature so that it is clear someone actually has deliberated the issue and taken a decision - ideally then and there.

Pass or fail!
ElMaestro
ssussu
☆    

China,
2018-02-08 05:23
(2240 d 07:59 ago)

@ ElMaestro
Posting: # 18378
Views: 9,175
 

 When the the infusion speed deviate from the protocol specified

Hi ElMaestro,

Thanks for your reminder. And I agree with you.
And I also think, investigator had to stop infusion to expel the bubble is for subject safety, but it occurs bubbles during infusion and this kind of things reflect the system quality, need to be improved and avoid.
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,639 registered users;
75 visitors (0 registered, 75 guests [including 3 identified bots]).
Forum time: 13:22 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5