0:01

I'm going to talk about principles of contrast

0:06

injection and these principles apply to all

0:11

vascular imaging and not just coronary CTA but

0:16

they're very important to understand because

0:19

that's how you get a great study like this one

0:23

on the left and you also need to understand why

0:26

they're sometimes bad, such as the one on the right.

0:30

Now ordinarily radiologists aren't

0:33

involved in this because this is entirely

0:36

taken care of by the technologists.

0:38

Protocols have been set up and you're kind

0:43

of only informed when things go wrong.

0:46

And, um, it's sort of like, you know,

0:48

the MD informed, you're informed, but what

0:49

are you going to do about it differently?

0:52

And to understand what to do differently,

0:54

you have to understand the principles as well

0:55

as the technologists and better than those.

0:58

So what is the fundamental

1:02

principle of contrast injection?

1:04

The fundamental principle is that.

1:07

You want to time the acquisition of CT, the picture

1:13

taking, to when the arteries are maximally opacified.

1:21

Now that sounds fairly simple,

1:26

except there are two things.

1:27

Firstly, when they're maximally opacified

1:31

arteries, it's only a short time, because

1:35

in arteries, contrast goes in and goes out.

1:40

Veins, they tend to hang out for a bit longer,

1:42

so veins are more forgiving than arteries.

1:45

The second thing is that different arterial

1:47

trees are opacified at different times.

1:51

Slightly different times, but getting optimality

1:54

means you need to think carefully about the

1:57

arterial tree that you're trying to image.

2:01

And the arterial tree, as in the case of the

2:05

coronary artery, is so small that for you to

2:08

actually time it, you'd have to see the coronary

2:12

arteries light up, and that can be quite difficult.

2:15

So what you do is you end up choosing the

2:17

artery closest to the arterial tree

2:20

of interest, which most resembles that.

2:24

And in the case of the coronary

2:25

artery, that happens to be the aorta.

2:28

And you can choose the descending

2:29

or the ascending aorta.

2:30

I don't think it makes that much of a difference.

2:32

And so now what you're doing is you're

2:33

trying to get the timing of the CT to

2:36

correspond with the peak maximal enhancement.

2:39

And that in itself is a bit of a science.

2:44

There are multiple factors to it.

2:46

There's timing, which I think is

2:49

what we have most control over.

2:52

There's breath hold.

2:53

Where, of course, you have control over, but

2:56

not that much, because ultimately it's patient

2:58

dependent, but you can educate the patient.

3:00

Bolus geometry, which is, again,

3:02

what we have control over.

3:04

And then there are patient factors, such as their

3:07

ejection fraction, which it's worth bearing in

3:10

mind, because you can do a variation on the theme.

3:14

And finally, there's the kVp, which I'll talk

3:17

about in more detail in another lecture, but,

3:19

yes, that's something to also bear in mind.

3:23

Peak maximum enhancement (PME) and time to peak (TTP)

3:28

are two important concepts which

3:31

happen behind the scenes, but they're happening

3:33

all the time and they depend on a bunch of factors.

3:37

Firstly, the rate of injection volume of contrast.

3:41

The amount of iodine in the contrast, so rate

3:44

volume iodine, so that's kind of like the iodine

3:46

flux, the amount of iodine that's going through.

3:49

And lastly, the saline push.

3:52

So if you look at the graph here, you'll notice

3:54

that as you increase the rate of contrast, if you

3:58

start at one milliliter per second, then you move on to

4:03

three milliliters and then five milliliters, the peak gets larger.

4:09

As you inject quicker, the peak

4:11

increases and the time to peak reduces.

4:15

So you get higher, quicker with

4:18

greater contrast injection.

4:20

That kind of makes sense.

4:23

This is an interesting one, the volume of contrast.

4:25

You'll notice that as the volume

4:26

increases, three things happen.

4:28

First is that your peak increases.

4:35

The second thing is your time to peak also increases.

4:39

It's kind of the opposite with the injection rate.

4:41

Injection rate peak increases.

4:43

Time to peak decreases here, the peak

4:45

increases, time to peak increases.

4:47

The third thing that happens, which may

4:49

not be best illustrated in this graph,

4:51

is that the time of that peak increases.

4:56

So instead of that pointy peak, it becomes

4:58

more like a nice little mountaintop.

5:03

So, basic contrast kinetics, also, if you're

5:08

increasing the amount of iodine, which you

5:11

don't really have an option of doing, because

5:12

you kind of get one iodinated agent, so you

5:15

can't say, let me put it in more contrast.

5:16

But something to bear in mind, if you

5:18

increase the amount of iodine, then it's

5:20

the same as increasing the injection rate.

5:23

So this graph here shows more iodine,

5:26

and then the blue shows less iodine.

5:29

More iodine, higher peak.

5:32

Quicker time to peak.

5:35

And then there are body factors as well,

5:37

but I think the most important one to understand

5:39

is this one here, which is what happens when

5:42

you have a reduction in the cardiac output.

5:45

So low ejection fraction cardiac failure situation

5:47

that you will frequently encounter.

5:50

And sadly, often only when something goes wrong.

5:54

So if you reduce your cardiac output, it's

5:58

the same as increasing the volume of contrast.

6:00

That's very interesting.

6:01

So it takes longer to get to the peak.

6:05

And the peak is higher.

6:06

And I think the peak is higher because your effective

6:09

circulating volume, because there's all these

6:12

kind of jazzy stuff that, uh, renal physiology and

6:15

pathophysiology, that what happens with heart failure

6:17

is even though you have more water on board, what's

6:20

actually in effective circulating volume reduces.

6:24

So you have hemoconcentration.

6:25

So that's the effect, same

6:27

effect as giving more volume.

6:28

So all of these things you want to bear in mind.

6:32

And also I didn't mention the saline push,

6:35

the saline push has the same effect, to an

6:37

extent, where it'll increase your time to peak.

6:43

And slightly increase your peak as well.

6:46

So in terms of timing, in terms of saying,

6:48

okay, when am I going to take the acquisition,

6:50

3T acquisition versus, you know, giving the

6:53

contrast, there are two ways you can do it.

6:55

You can do an empirical delay, you can make a

6:56

guesswork, and this is what we do with veins.

6:59

So, you know, we kind of say with a portal

7:00

vein, 70 seconds, the lower extremity veins,

7:03

120 to 140 seconds, or you can trigger.

7:06

Trigger is when you try and be precise about it.

7:10

There are two ways to trigger, one is called bolus

7:12

tracking and the other one is called test bolus.

7:13

I'll go through test bolus because that's

7:15

used less often than bolus tracking.

7:19

So in test bolus what you're doing is you're

7:20

injecting a small amount of contrast, let's say

7:22

about 20 mL or so, in a predetermined segment,

7:26

such as the aorta, and then you're seeing

7:28

how long does it take for it to reach the peak.

7:33

The peak itself isn't important, but

7:36

it's the time to peak that's important.

7:37

And then however long it takes to get to that peak.

7:44

With the real injection, which is going to be much

7:47

more, I factor in a fudge factor of about five to

7:50

six seconds, because if you recall, when you give

7:53

more contrast, it takes longer to get to the peak.

7:58

So it's a method of, um, calibrating.

8:03

When do I use it?

8:04

I probably should use it more often,

8:06

but I a hundred percent use it if I know

8:09

the patient has low ejection fraction.

8:13

So I do get the texts for

8:16

you know, things like runoffs and,

8:18

aortic dissection studies and even

8:21

coronaries to see if their ejection fraction is low.

8:24

So how low is low?

8:26

I mean, this is an arbitrary number I'm putting

8:28

out, but you know, less than 20%, absolutely.

8:32

Yeah.

8:32

20 to 30%.

8:33

You might get away with the other method,

8:36

but yes, less than 20 percent clearly.

8:41

All right.

8:41

So this is the method we use all the

8:43

time, which is the bolus tracking method.

8:50

In the bolus tracking method, what we do is we

8:55

put a tracker on the abdominal aorta or somewhere

9:02

and watch how it increases in signal intensity.

9:07

We specify a threshold such as 130 Hounsfield

9:10

units and we then specify a trigger delay.

9:17

So, the trigger delay is the time from

9:20

reaching the Hounsfield unit threshold

9:23

to the start of the acquisition.

9:25

You do need to have some trigger delay,

9:27

because if you don't, you're going to end up

9:29

with, I mean, it's not possible that there

9:31

has to be this deep breath in and hold it.

9:36

So I want you to look at this, because it's a very

9:40

busy graph, and I'm going to go through this again.

9:44

But an important point to understand over

9:48

here is that when you give the injection,

9:51

you don't take the images straight away.

9:54

Because when you give the injection, you

9:56

give it through a vein, it goes through your

9:58

circulation, goes through the right heart, comes

10:01

to the pulmonary artery, then pulmonary vein,

10:03

then left atrium, left ventricle, then aorta.

10:06

And then you do what's called a monitoring.

10:08

You start monitoring the arrival of

10:11

contrast in that predetermined segment.

10:13

When it reaches that threshold, then after that

10:16

trigger delay, you start the scan with the hope of

10:21

coinciding the acquisition with the peak maximum

10:24

enhancement. It's important to understand that the

10:27

actual scan happens after 15 to 20 seconds towards

10:33

the tail end of the injection, and that's why your

10:36

injection can't be too short, because you'll miss that.

10:41

There are two types of delay.

10:42

There is the monitoring delay, which

10:44

is the delay between the start of the

10:45

injection and the first monitoring slice.

10:48

You have to use that judiciously, but the

10:50

trigger delay is the one that's different.

10:53

So you have to, when communicating to techs, make

10:55

sure you understand the distinction between trigger

10:57

delay, which has been in the attenuation, reaching

11:00

a preset threshold and the start of the acquisition.