I was was trained (June Boon) to use the right parasternal 5 chamber view for measuring IVS/LV/PW in m-mode but I see that also the 4 chamber view is recommended. So which is better?
I was was trained (June Boon) to use the right parasternal 5 chamber view for measuring IVS/LV/PW in m-mode but I see that also the 4 chamber view is recommended. So which is better?
Comments
4 vs 5 chamber view for m-mode
I’ve seen it done both ways and I don;t think there is true consensus but Peter may have more insight. The 5 chamber view in my opinion risks measuring too much of the fibrous part of the ivs that is stiffer than the true muscularis part of the LV septum but this is subjective. Some number crunching researchers would need to study this. Peter any insight here?
4 vs 5 chamber view for m-mode
I’ve seen it done both ways and I don;t think there is true consensus but Peter may have more insight. The 5 chamber view in my opinion risks measuring too much of the fibrous part of the ivs that is stiffer than the true muscularis part of the LV septum but this is subjective. Some number crunching researchers would need to study this. Peter any insight here?
Yes, unfortunately theres no
Yes, unfortunately theres no consensus
i know that most US cardiologists use 5 chamber views or Short axis views. In our College of cardiology we agreed using 4 chamber views or short axis views. I got the impression that 4 chamber views are higher reproducible (repeatability, Inter- observer) than 5 chamber views. And its very simple to obtain the same results as With short axis views.
BTW I dont think that many cardiologists use June Boons reference tables any more. Most use Cornell’s indices ( Cornell et at, JVIM 2004).
Summarizing, it’s ok to use 5 chamber views but I recommend using 4 chamber views or Short axis views
Best regards from the Austrian Mountains
Peter
Yes, unfortunately theres no
Yes, unfortunately theres no consensus
i know that most US cardiologists use 5 chamber views or Short axis views. In our College of cardiology we agreed using 4 chamber views or short axis views. I got the impression that 4 chamber views are higher reproducible (repeatability, Inter- observer) than 5 chamber views. And its very simple to obtain the same results as With short axis views.
BTW I dont think that many cardiologists use June Boons reference tables any more. Most use Cornell’s indices ( Cornell et at, JVIM 2004).
Summarizing, it’s ok to use 5 chamber views but I recommend using 4 chamber views or Short axis views
Best regards from the Austrian Mountains
Peter
Thanks – good to know.I have
Thanks – good to know.
I have switched over to using the Cornell tables as well. (well I look at both actually) I am finding that I am over-diagnosing things using June’s tables more often than not. There is a really nice echo parameter calculator on VIN for this on the cardiology board that is easy to use. I do have some excel spread sheets to do the calculations as well but they are a little confusing and I seem to mess them up too easily.
Thanks – good to know.I have
Thanks – good to know.
I have switched over to using the Cornell tables as well. (well I look at both actually) I am finding that I am over-diagnosing things using June’s tables more often than not. There is a really nice echo parameter calculator on VIN for this on the cardiology board that is easy to use. I do have some excel spread sheets to do the calculations as well but they are a little confusing and I seem to mess them up too easily.
I wanted to add a comment
I wanted to add a comment regarding 4 chamber vs. 5 chamber measurements.
For those of us that use the cornell/kittleson calculator (on VIN or spreadsheet) , I am thinking that the input data should be based on m-mode measurements taken from the right long axis parasternal (5-chamber “inflow-outflow” view ) OR right transverse parasternal view and not 4 chamber. I say this because the JVIM paper that describes the formulation of the cornell/kittleson equation is based on data collected from retrospective studies by several sonographers using m-mode in either the right parasternal 5 chamber or transverse view. So I would be concerned measurements taken from 4 chamber views may not give a valid assessment of the dog’s heart using this equation. I am not saying one imaging view is better than the other, just that we should feed the right data into the cornell/kittleson equation for the best result.
I wanted to add a comment
I wanted to add a comment regarding 4 chamber vs. 5 chamber measurements.
For those of us that use the cornell/kittleson calculator (on VIN or spreadsheet) , I am thinking that the input data should be based on m-mode measurements taken from the right long axis parasternal (5-chamber “inflow-outflow” view ) OR right transverse parasternal view and not 4 chamber. I say this because the JVIM paper that describes the formulation of the cornell/kittleson equation is based on data collected from retrospective studies by several sonographers using m-mode in either the right parasternal 5 chamber or transverse view. So I would be concerned measurements taken from 4 chamber views may not give a valid assessment of the dog’s heart using this equation. I am not saying one imaging view is better than the other, just that we should feed the right data into the cornell/kittleson equation for the best result.
4 vs 5 chamber for view m-mode
Hi Tom, thanks for your important comment. Yes, you should always use the same method when using reference interval.
Still, I have to mention the following:
In Cornell’s paper there’s no description of how M-Mode measurements of the left ventricle were done (at least I did not find it). The vast majority of the data included were unpublished personal data from the different contributors. But: As I know some of the authors and their data (and way of data acquisition) very well, I’m totally convinced that at least a significant part of the M-Mode studies included were obtained from 4-chamber views. Thus, we do not know, whether the equation is better for 4- or 5-chamber M-Modes. But given a CV of ~12-15% for M-Mode measurements, I would say that it’s very unlikely that a difference between M-Modes obtained from 5- or 4-chamber views matter (given that there is one…)
Summarizing: Feed the formula with whatever you like.
Since I’m not on VIN: Which prediction interval do they use? (I prefer the 95%)
best regards
Peter
4 vs 5 chamber for view m-mode
Hi Tom, thanks for your important comment. Yes, you should always use the same method when using reference interval.
Still, I have to mention the following:
In Cornell’s paper there’s no description of how M-Mode measurements of the left ventricle were done (at least I did not find it). The vast majority of the data included were unpublished personal data from the different contributors. But: As I know some of the authors and their data (and way of data acquisition) very well, I’m totally convinced that at least a significant part of the M-Mode studies included were obtained from 4-chamber views. Thus, we do not know, whether the equation is better for 4- or 5-chamber M-Modes. But given a CV of ~12-15% for M-Mode measurements, I would say that it’s very unlikely that a difference between M-Modes obtained from 5- or 4-chamber views matter (given that there is one…)
Summarizing: Feed the formula with whatever you like.
Since I’m not on VIN: Which prediction interval do they use? (I prefer the 95%)
best regards
Peter
… But I still prefer the
… But I still prefer the 4-chamber or short axis views… 🙂
… But I still prefer the
… But I still prefer the 4-chamber or short axis views… 🙂
Peter,You are correct.
Peter,
You are correct. Cornell’s paper does not specify a view in which the M-Mode measurements of the left ventricle were obtained in the 12 retrospective studies. Apparently there was more than one technique used. Which just confirms the “no consensus” comments that you and Eric have made.
On VIN, the website it explains: “Two reference intervals are available for this calculation (Cornell/Kittleson) method – one for less experienced echocardiographers, which takes the mean calculated value +/- 20% to set the reference interval, and one for more experienced echocardiographers, which takes the mean calculated value +/- 10% to set the reference interval.”
I have since emailed a VIN cardiologist about the use of the Cornell/Kittleson formula available on VIN. He prefers to make LV m-mode measurements from the four chamber view! Even though, the schematic on VIN’s calculator site, demonstrates the LV chamber and wall measurements in the long axis inflow/outflow view (not 4 chamber) !
Like Pankatz, I have been taught to measure from the right parasternal inflow-outflow long axis view or short axis view on the same side. I plug those numbers in a spreadsheet that calculates a 95% CI using the Cornell equation.
Given your (and Eric’s) preference, I will give measuring from the 4 chamber view a try. This seems to be an easier view to capture in the dog. But I will likely stay with what I have been taught. Old habits are hard to break:)
Appreciated your comments and the discussion in general!
Tom
Peter,You are correct.
Peter,
You are correct. Cornell’s paper does not specify a view in which the M-Mode measurements of the left ventricle were obtained in the 12 retrospective studies. Apparently there was more than one technique used. Which just confirms the “no consensus” comments that you and Eric have made.
On VIN, the website it explains: “Two reference intervals are available for this calculation (Cornell/Kittleson) method – one for less experienced echocardiographers, which takes the mean calculated value +/- 20% to set the reference interval, and one for more experienced echocardiographers, which takes the mean calculated value +/- 10% to set the reference interval.”
I have since emailed a VIN cardiologist about the use of the Cornell/Kittleson formula available on VIN. He prefers to make LV m-mode measurements from the four chamber view! Even though, the schematic on VIN’s calculator site, demonstrates the LV chamber and wall measurements in the long axis inflow/outflow view (not 4 chamber) !
Like Pankatz, I have been taught to measure from the right parasternal inflow-outflow long axis view or short axis view on the same side. I plug those numbers in a spreadsheet that calculates a 95% CI using the Cornell equation.
Given your (and Eric’s) preference, I will give measuring from the 4 chamber view a try. This seems to be an easier view to capture in the dog. But I will likely stay with what I have been taught. Old habits are hard to break:)
Appreciated your comments and the discussion in general!
Tom
Peter,You are correct.
Peter,
You are correct. Cornell’s paper does not specify a view in which the M-Mode measurements of the left ventricle were obtained in the 12 retrospective studies. Apparently there was more than one technique used. Which just confirms the “no consensus” comments that you and Eric have made.
On VIN, the website it explains: “Two reference intervals are available for this calculation (Cornell/Kittleson) method – one for less experienced echocardiographers, which takes the mean calculated value +/- 20% to set the reference interval, and one for more experienced echocardiographers, which takes the mean calculated value +/- 10% to set the reference interval.”
I have since emailed a VIN cardiologist about the use of the Cornell/Kittleson formula available on VIN. He prefers to make LV m-mode measurements from the four chamber view! Even though, the schematic on VIN’s calculator site, demonstrates the LV chamber and wall measurements in the long axis inflow/outflow view (not 4 chamber) !
Like Pankatz, I have been taught to measure from the right parasternal inflow-outflow long axis view or short axis view on the same side. I plug those numbers in a spreadsheet that calculates a 95% CI using the Cornell equation.
Given your (and Eric’s) preference, I will give measuring from the 4 chamber view a try. This seems to be an easier view to capture in the dog. But I will likely stay with what I have been taught. Old habits are hard to break:)
Appreciated your comments and the discussion in general!
Tom
Peter,You are correct.
Peter,
You are correct. Cornell’s paper does not specify a view in which the M-Mode measurements of the left ventricle were obtained in the 12 retrospective studies. Apparently there was more than one technique used. Which just confirms the “no consensus” comments that you and Eric have made.
On VIN, the website it explains: “Two reference intervals are available for this calculation (Cornell/Kittleson) method – one for less experienced echocardiographers, which takes the mean calculated value +/- 20% to set the reference interval, and one for more experienced echocardiographers, which takes the mean calculated value +/- 10% to set the reference interval.”
I have since emailed a VIN cardiologist about the use of the Cornell/Kittleson formula available on VIN. He prefers to make LV m-mode measurements from the four chamber view! Even though, the schematic on VIN’s calculator site, demonstrates the LV chamber and wall measurements in the long axis inflow/outflow view (not 4 chamber) !
Like Pankatz, I have been taught to measure from the right parasternal inflow-outflow long axis view or short axis view on the same side. I plug those numbers in a spreadsheet that calculates a 95% CI using the Cornell equation.
Given your (and Eric’s) preference, I will give measuring from the 4 chamber view a try. This seems to be an easier view to capture in the dog. But I will likely stay with what I have been taught. Old habits are hard to break:)
Appreciated your comments and the discussion in general!
Tom
Hi Tom!Many thanks for your
Hi Tom!
Many thanks for your interesting comments and the information about the VIN calculator! This discussion shows clearly the limitation of all studies presenting reference values. If you look through the literature, you will find that the way M-Modes were obtained is not specified in the vast majority of the studies. Furthermore, almost (if not all) of the studies were cross-sectional, thus, no one knows, if these dogs developed e.g. DCM 1 or 2 years later. And: To provide reference values for a specific breed, you would need at least 100 individuals to achieve an appropriate statistical power.
It has also to be mentioned that different breeds (particularly large breeds as opposed to small breeds ) can have largely different M-Mode values. Putting all of them together into a study is a major limitation (Breeds do not differ only by body weight!). The same for the effect of different loading conditions. And the inclusion of racing dogs a.s.o.
In most cases (except racing dogs, except dogs weighing more than 40 kg), the following pragmatic rule applies (it’s my own one I use for teaching, not published, no consensus, no proof):
Looking at a right parasternal long axis 4 chamber view, the heart is hemodynamically normal (no volume overload, no pressure overload; of course some minor stenosis or insufficiency whcih is hemodynamically irrelevant can be present…) if:
-The IVS is straight, the IAS is straight, the LVW fits 3.5-4.5 times into the LV at the end of diastole (look at Cornell’s table No 4 up to 40 kg, it seems to apply his population as well).
-The left atrium is a cube and fits approx 2x into the LV at end-diastole
– The right free wall is 1/2 of the septum or LVw at diastole.
– The right ventricle is 1/2-1/3 of the LV at end-diastole
– the right atrium is smaller than the left one
For breeding exams, I use the published reference values to be on the safe side. They are particularly helpful in specific breeds (Doberman – detailled values provided by Gerhard Wess, I Wolf – Andrea Vollmar, Great Danes,…). When the FS is at the lower end (20-25%) and I do not find a specific reason for that (not an athletic dog, not dehydrated,…), I calculate EF by use of Simpson’s method or do some tissue Doppler measurements to get more information. Simpson is particularly helpful for Dobermans (95 ml/m2, 55 ml/m2) – but only for European Dobermans?????
One major reason why I do not use 5 chamber views for M-Modes is that the 5-chamber view is much more variable (visibility of mitral valve leaflets – not only the left atrium should be displayed; tipped or horizontal views, length of the left ventricle…), the 4 chamber view is much more consistent and repeatable. But, as already mentioned: A highly trained sonographer (and that’s what I heard about you) can get optimal M-Modes in any of the views…
Best regards and thanks for the discussion!
Peter
Hi Tom!Many thanks for your
Hi Tom!
Many thanks for your interesting comments and the information about the VIN calculator! This discussion shows clearly the limitation of all studies presenting reference values. If you look through the literature, you will find that the way M-Modes were obtained is not specified in the vast majority of the studies. Furthermore, almost (if not all) of the studies were cross-sectional, thus, no one knows, if these dogs developed e.g. DCM 1 or 2 years later. And: To provide reference values for a specific breed, you would need at least 100 individuals to achieve an appropriate statistical power.
It has also to be mentioned that different breeds (particularly large breeds as opposed to small breeds ) can have largely different M-Mode values. Putting all of them together into a study is a major limitation (Breeds do not differ only by body weight!). The same for the effect of different loading conditions. And the inclusion of racing dogs a.s.o.
In most cases (except racing dogs, except dogs weighing more than 40 kg), the following pragmatic rule applies (it’s my own one I use for teaching, not published, no consensus, no proof):
Looking at a right parasternal long axis 4 chamber view, the heart is hemodynamically normal (no volume overload, no pressure overload; of course some minor stenosis or insufficiency whcih is hemodynamically irrelevant can be present…) if:
-The IVS is straight, the IAS is straight, the LVW fits 3.5-4.5 times into the LV at the end of diastole (look at Cornell’s table No 4 up to 40 kg, it seems to apply his population as well).
-The left atrium is a cube and fits approx 2x into the LV at end-diastole
– The right free wall is 1/2 of the septum or LVw at diastole.
– The right ventricle is 1/2-1/3 of the LV at end-diastole
– the right atrium is smaller than the left one
For breeding exams, I use the published reference values to be on the safe side. They are particularly helpful in specific breeds (Doberman – detailled values provided by Gerhard Wess, I Wolf – Andrea Vollmar, Great Danes,…). When the FS is at the lower end (20-25%) and I do not find a specific reason for that (not an athletic dog, not dehydrated,…), I calculate EF by use of Simpson’s method or do some tissue Doppler measurements to get more information. Simpson is particularly helpful for Dobermans (95 ml/m2, 55 ml/m2) – but only for European Dobermans?????
One major reason why I do not use 5 chamber views for M-Modes is that the 5-chamber view is much more variable (visibility of mitral valve leaflets – not only the left atrium should be displayed; tipped or horizontal views, length of the left ventricle…), the 4 chamber view is much more consistent and repeatable. But, as already mentioned: A highly trained sonographer (and that’s what I heard about you) can get optimal M-Modes in any of the views…
Best regards and thanks for the discussion!
Peter
Sorry to open this discussion
Sorry to open this discussion again! But for those of you using the 4-chamber for the m-mode measurements, are you also using it to measure EPSS or is this only for the LV walls and chamber size?
Sorry to open this discussion
Sorry to open this discussion again! But for those of you using the 4-chamber for the m-mode measurements, are you also using it to measure EPSS or is this only for the LV walls and chamber size?