Perfusion data entry

[GaelleChapuis]

The histology WG noted that there is some variability in the auto-fluorescence brightness signal. There seems to be some unknown cause of protocol discrepancy. To help us tackling this, we would like to add some fields, notably:

• The PFA solution
• The goodness of the perfusion (how to assess and label is under debate)
• Transport time + temperature (e.g. room temperature or on ice)
• Time from perfusion to microscopy imaging (can be computed using death date and 'imaging date', if existing in the database upon histology session loading)

Exact implementation to be discussed.

[anne-urai]

Time from first craniotomy to perfusion (overall brain health/infection)

[lhaetzel]

Regarding the quality of the perfusion, some tell-tale signs of a poor perfusion are: bleeding from the nose during perfusion, heart stopped beating before needle insertion, liver does not lose red color during perfusion (blanching)

[GaelleChapuis]

After talking with Steven, a more thorough list is:

• The PFA solution : IBL receipe v0.0.1 ; other
• The goodness of the perfusion - liver cleared : yes / no (boolean)
• Post-fixation time (in hours)
• Post-fixation temperature : 4°C ; Room temp. ; other
• Transport time (in hours)
• Transport temperature : 4°C ; On dry ice ; Room temp. ; other
• Transport solution : 30% sucrose PBS ; PBS ; other
• Time from perfusion to microscopy imaging (in hours)

fields with other are machine readable ; indicate differences in perfusion notes

[GaelleChapuis]

@anne-urai , I did not add ''Time from first craniotomy to perfusion'' as this can be already retrievable from Alyx via the different surgery dates. Would it be simpler for analysis to have a field with this information in though (could be done automatically @rossant ?)?

[stevenjwest]

Hi all, I just wanted to copy-paste some of my discussion with @GaelleChapuis here for you all to see:

I believe the best way to assess perfusion is by confirming the liver has cleared of blood during the flushing with PBS. This provides a convenient and clear (no pun intended!) classification between a 'good' and 'poor' perfusion. Everyone should really be using the EM grade 16% Formaldehyde vials I listed on the Perfusion SOP, as the formaldehyde made from these will be very pure, and will not contain any small particles of PFA, that often remain when you depolymerise PFA to make formaldehyde solution, which can occlude small blood vessels during perfusion (there is a paper on this, but I cannot find it right now).

The perfusion SOP currently lists a standard of 24 hours at 4C, although I think longer would not be a problem. Under-fixation could be a problem though, and it would be good to check the fixation time to see if it makes any impact.

In terms of assessing perfusion quality, in my experience the perfusion is either good, and the liver will clear of blood, or if you mess it up, the liver just doesn't clear. As the quality of perfusion really is determined by how good a fit your needle makes with the heart ventricle - if you puncture the heart so your perfusate flows out of that hole and not through the circulatory system, you have a bad perfusion. Otherwise it "should" be good.

If perfusate starts to come out through the nose, this could potentially indicate a bad perfusion, although that does vary somewhat, as this just means the needle has punctured the wall between left and right ventricles, and now is pumping perfusate through both the left and right ventricles - with perfusate going through the right ventricle heading through the lungs and then eventually being released via the nose. But if this does happen, often I find the perfusion is still OK.

I would also just like to add that when I do a perfusion I actually cull before beginning the perfusion with an overdose of isoflurane, which I deem as more humane. In this state, the heart is in fibrillation, which makes it a lot easier to setup the perfusion, and I still get absolutely fine perfusions and subsequent histology. So it isn't necessary for the heart to be beating - this is a myth that seems to perpetuate in histology circles. I should also say I have used this method for perfusions for the past 5-6 years, and I only do histology, so it has been well validated!

I have incorporated these notes into the Perfusion SOP.

[GaelleChapuis]

@rossant when could this be implemented?

[rossant]

Here's a proposition:

class Perfusion(BaseAction):
    pfa_solution = models.ForeignKey(
        PerfusionSolution, null=True, blank=True, on_delete=models.SET_NULL,
        related_name='pfa_solution')

    liver_cleared = models.BooleanField(blank=True, default=False)

    post_fixation_time = models.PositiveIntegerField(blank=True, default=0)

    post_fixation_temperature = models.ForeignKey(
        PerfusionTemperature, null=True, blank=True, on_delete=models.SET_NULL,
        related_name='post_fixation_temperature')

    transport_time = models.PositiveIntegerField(blank=True, default=0)

    transport_temperature = models.ForeignKey(
        PerfusionTemperature, null=True, blank=True, on_delete=models.SET_NULL,
        related_name='transport_temperature')

    transport_solution = models.ForeignKey(
        PerfusionSolution, null=True, blank=True, on_delete=models.SET_NULL,
        related_name='transport_solution')

    time_to_imaging = models.PositiveIntegerField(blank=True, default=0)

As a BaseAction, this new model will also have the following fields:

users
subject
location
lab
procedures
narrative
start_time
end_time

There will be initial fixtures, as follows:

Perfusion temperature:

Room temperature 
4°C 
Dry ice
Other 


Perfusion solution:

IBL receipe v0.0.1
30% sucrose PBS
PBS
Other

Is that correct?

[GaelleChapuis]

1 .---------------------------------------------------------------

Based on feedback, I would suggest making an object '' Solution change '' that contains the fields:

• Date/Time at which the solution is changed:
• New solution type:
• Storage temperature:

---

Fields for solution type:

• PBS
• PBS 30% sucrose
• PFA IBL solution v0.0.1
• Other

That object could be added as needed. Users would input to start with: (example)

• Date/time: (pre-define as today's date) ; time: (pre-define as today's date)
• Solution: PFA
• Storage temperature: 4degC

Then upon changing into PBS or transport solution: (example)

• Date/time: (pre-define as today's date) ; time: (pre-define as today's date)
• Solution: PBS with 30% sucrose
• Storage temperature: Dry ice

Remove field:

• post-fixation time

Edit field:

• date of perfusion (written now as start time - just have the date (pre-define as today's date), specify it's perfusion ; time will come from the solution object mentioned above)

2. -----------------------------------------------------
Make sure 'Other' is at the bottom of drop down menu.

3.-----------------------------------------------------

Remove temperature of transport, and put instead:

• Type of transport: By airmail ; In person; Other
• Date of transport (sending date) : pre-define to today's date

4.-----------------------------------------------------
Remove time to imaging, and add in another histology session schema:

• date of receiving the sample
• date of imaging