IM intern vs. RT: ABG smackdown

[Redpancreas]

We’ve all been there on wards in a nonintubated patient. The patient is being admitted for hypoxia in ER/Wards with SaO2s reading somewhere in the 90s with occasional dips into the high 80s is reporting shortness of breath. The RN throws on 2L and call the RT who proceeds to argue against an ABG but you have to argue for it because your attending wants it.

I can understand why pH, and even CO2 are important but those are on the VBG. Why do you need the PaO2 though? I know you can calculate the A-a gradient and that can change your differential. If SaO2 can just be converted to PaO2 though, why not just do that? Why do we get an ABG when we have continuous SaO2 (hence PaO2) monitoring?

I understand there may be a very good reason for this and this may be a stupid question but I can’t seem to find the answer. Thanks!

---

Members don't see this ad.

 

[DrMetal]

We’ve all been there on wards in a nonintubated patient. The patient is being admitted for hypoxia in ER/Wards with SaO2s reading somewhere in the 90s with occasional dips into the high 80s is reporting shortness of breath. The RN throws on 2L and call the RT who proceeds to argue against an ABG but you have to argue for it because your attending wants it.

I can understand why pH, and even CO2 are important but those are on the VBG. Why do you need the PaO2 though? I know you can calculate the A-a gradient and that can change your differential. If SaO2 can just be converted to PaO2 though, why not just do that? Why do we get an ABG when we have continuous SaO2 (hence PaO2) monitoring?

I understand there may be a very good reason for this and this may be a stupid question but I can’t seem to find the answer. Thanks!

Did you mean continuous SpO2? We get the ABG when we want a more accurate number.

 

[NewYorkDoctors]

We’ve all been there on wards in a nonintubated patient. The patient is being admitted for hypoxia in ER/Wards with SaO2s reading somewhere in the 90s with occasional dips into the high 80s is reporting shortness of breath. The RN throws on 2L and call the RT who proceeds to argue against an ABG but you have to argue for it because your attending wants it.

I can understand why pH, and even CO2 are important but those are on the VBG. Why do you need the PaO2 though? I know you can calculate the A-a gradient and that can change your differential. If SaO2 can just be converted to PaO2 though, why not just do that? Why do we get an ABG when we have continuous SaO2 (hence PaO2) monitoring?

I understand there may be a very good reason for this and this may be a stupid question but I can’t seem to find the answer. Thanks!

to be fair, RTs tell me they only do ROUTINE scheduled ABGs. they are not the STAT ABG person. from a logistic standpoint that would make some sense as they have a lot of vents to manage, PAPs to manage, suctions, nebulizers etc...

the ICU RTs are usually more cool about it and will do the ABG on demand since that plays into extubation possibly.


honestly the PaO2 is not terribly useful during a rapid response situation. the only situation is if the SpO2 is not reliable or not picking up (and the forehead or ear lobe oximeters are not giving any useful information)

PaO2 is quite useful outpatient during a CPET if ABGs are drawn for dead space calculations and A-a gradient, testing before and after 100% FiO2 to determine V/Q mismatch vs shunt physiology... but that's all draw nout slower paced chronic workup

But the precise A-a gradient is never so important (meaning it does not make or break the imminent rapid response patient's management in the short term) that it MUST be obtained for the sake of knowing the PaO2.
I mean its nice for PCP pneumonia but again that steroid decision is not something to be made on the spot at the RRT

for ARDS doing a formal PaO2 / FiO2 ratio is useful for the Berlin criteria but it does not change any of the clinical management as it comes to sedation, paralysis, low tidal volumes, permissive hypercapnia, proning, etc..
but it might signify the need for ECMO perhaps. so that's another common situation where PaO2 is helpful in the ICU

bottom line you really want the pH and PaCO2 and then the BMP for HCO3, K... the acute stuff.

PaO2 really may help is if there is any clinical status change that causes a LEFT SHIFT of the hemoglobin dissociation curve
as you see if left shifted, then for a given SpO2 measurement, there is actually a far lower PaO2 value and one might be lulled to sleep by the comfort of a normal looking SpO2 value.

Another situation in which getting a Co-Ox profile (same ABG machine) is if you are suspecting methemoglobinemia (classically 85% fixed on SpO2 per boards... though that's not always the case), carbohyhemoglobinemia (SpO2 is normal due to the CoHb making the infrared light read high but there is something wrong with the electron transport chain...), or sulfhemoglobinemia (the sewer history and green blood tends to give that away)

ultimately for the crashing patient on the floors, waiting on a PaO2 value may not changing the decision for intubation. end of the day if the patient is in extremis and desatting, nothing wrong with intubation shipping to ICU (the fellow may mutter but fellows don't get paid $ yet so they always mutter), then making sure patient is okay and shipping back to floors ASAP.

this is better than second guessing, putting someone on BiPAP and then they crash out.

 

[end stage fibro]

Excellent post above.

I’m several years out from residency and have lowered my ABG ordering substantially. A good pleth and SpO2 seems plenty accurate to me and gives some sense about perfusion. I only get ABGs now if i am worried about CO2 or pH.

One facility I work at, the lab draws the VBG. It’s incredible…no pushback or anything just a phleb tech at the bedside in minutes drawing blood.

 

[NewYorkDoctors]

Excellent post above.

I’m several years out from residency and have lowered my ABG ordering substantially. A good pleth and SpO2 seems plenty accurate to me and gives some sense about perfusion. I only get ABGs now if i am worried about CO2 or pH.

One facility I work at, the lab draws the VBG. It’s incredible…no pushback or anything just a phleb tech at the bedside in minutes drawing blood.

occassionally a phlebotomy tech finds a nice vessel at the lateral aspect of the wrist (the presumed cephalic vein)
but it was actually a superficial palmar arch of the radial artery.
then i see the PO2 is high. high five free ABG from the phlebotomy tech

 

[ShockIndex]

PaO2 is quite useful outpatient during a CPET if ABGs are drawn for dead space calculations and A-a gradient, testing before and after 100% FiO2 to determine V/Q mismatch vs shunt physiology... but that's all draw nout slower paced chronic workup

for ARDS doing a formal PaO2 / FiO2 ratio is useful for the Berlin criteria but it does not change any of the clinical management as it comes to sedation, paralysis, low tidal volumes, permissive hypercapnia, proning, etc..
but it might signify the need for ECMO perhaps. so that's another common situation where PaO2 is helpful in the ICU

I’m an intensivist who doesn’t do pulm or CPET, so you may have me at a disadvantage. However, I’m not sure how a PaO2 can be used to calculate dead space. I suspect that you meant PaCO2 is being used to determine dead space.

PaO2 can be used to calculate shunt fraction according to the Berggren equation if you know the mixed venous and pulmonary end-capillaries O2 content. In other words, such calculations are difficult and invasive. This is greatly simplified by using a P/F to estimate the severity of shunt physiology.

Dead space is going to be estimated by PaCO2 and end-tidal CO2 according to the Boher equation. This is nothing more than the fractional difference between PaCO2 and ETCO2. In other words, that patient who has been on a vent with COVID for a month who has a PaCO2 of 70 mmHg and an ETCO2 of 40 mmHg has a dead space fraction of almost 0.45 which is why their minute ventilation is 12LPM - almost half that is dead space.

Let me know if I’m missing something with the CPET calculations that may be different than what I use in the unit. Otherwise, I think you’re right on point. ABGs are often unnecessary to appropriately manage a patient if you have a normal VBG pH and PCO2 as well as a good pleth SpO2.

 

[NewYorkDoctors]

I’m an intensivist who doesn’t do pulm or CPET, so you may have me at a disadvantage. However, I’m not sure how a PaO2 can be used to calculate dead space. I suspect that you meant PaCO2 is being used to determine dead space.

PaO2 can be used to calculate shunt fraction according to the Berggren equation if you know the mixed venous and pulmonary end-capillaries O2 content. In other words, such calculations are difficult and invasive. This is greatly simplified by using a P/F to estimate the severity of shunt physiology.

Dead space is going to be estimated by PaCO2 and end-tidal CO2 according to the Boher equation. This is nothing more than the fractional difference between PaCO2 and ETCO2. In other words, that patient who has been on a vent with COVID for a month who has a PaCO2 of 70 mmHg and an ETCO2 of 40 mmHg has a dead space fraction of almost 0.45 which is why their minute ventilation is 12LPM - almost half that is dead space.

Let me know if I’m missing something with the CPET calculations that may be different than what I use in the unit. Otherwise, I think you’re right on point. ABGs are often unnecessary to appropriately manage a patient if you have a normal VBG pH and PCO2 as well as a good pleth SpO2.

You are correct . I just mistyped in my phone and got a lot of things conflated together while mashing my fingers crossed rapidly . Thank goodness for peer review .

There is a role for measuring A-a gradient and changes in Vd / Vt [(PaCO2 - PECO2 )/PaCO2) before and after exercise in certain cases of exercise induced PH and possible exercise induced R to L shunt.

But I’ve found most common cases are done just fine with the noninvasive cpet . It yields quite a good insight into the mechanisms of dyspnea .

 

[gulmohr]

 

[NewYorkDoctors]

View attachment 375801

nice. i havent done any full time ICU work since COVID 2020 at the start so thanks for the update