The Critical Care Advice Column: AMA

Brandon O

Puzzled by facies
1,718
337
83
Hi all,

This is a open thread for anybody with questions about critical care. Have a patient scenario to discuss? Confused about a drug or some physiological mechanism? I guarantee I will answer anything to the best of my ability. (Of course, others are welcome to chime in too!)

My only caveat is that I may "steal" some selected questions to answer on my website. (I'll leave the asker anonymous if desired.)

For a bit of reference, I'm a residency-trained critical care PA with a background in EMS.

Anyone got a burning question about sick patients?
 

VentMonkey

Family Guy
5,729
5,043
113
What can you tell me about driving pressures and their importance with ARDS patients?

I want to say I was reading recently that they are thought to play a more significant role in vent management of the ARDS patient, along with obviously lower Vt's.
 
OP
OP
Brandon O

Brandon O

Puzzled by facies
1,718
337
83
What can you tell me about driving pressures and their importance with ARDS patients?

I want to say I was reading recently that they are thought play a more significant role in vent management of the ARDS patient, along with obviously lower Vt's.

So the best-known and best-proven approach to "lung-protective" ventilation for ARDS patients is to limit plateau pressures and tidal volumes, which is proven to improve mortality. The idea is that an ARDS lung has reduced compliance, due to both interstitial changes (lung is less stretchy) and airspace disease (there's less actual lung to ventilate, since some of it is filled with crud). Forcing high pressures/volumes into that causes damage, aka volutrauma.

So we measure the volumes and/or pressures we give. Both are really surrogate measures, though, since we don't know how much lung we're actually ventilating and we don't know how far we can stretch those alveolar before injuring them. Thus we're always looking for better metrics.

Driving pressure was a proposed alternative. It was most popularized by Amato 2015, which reanalyzed the original ARDSnet data looking for a different number called "driving pressure." This is essentially plateau pressure minus the PEEP. Since plateau pressure is the pressure in the alveolar at the end of inspiration, and PEEP is the pressure at the start of inspiration, driving pressure essentially reflects the actual pressure/volume change (delta P) from the breath. Arguably this would better reflect lung protection than plateau alone, which includes PEEP even though the "PEEP air" isn't moving.

You would perform a "driving pressure trial" something like this:
  1. Give a test breath, check driving pressure. Let's say PEEP is 5, plateau 20, so driving pressure is 15.
  2. Increase PEEP a little, check driving pressure after a little while. Let's say PEEP is 7, plateau is now 21. Driving pressure is now LESS at 14. What does this mean? It means you recruited more lung by increasing PEEP. There's more space to put the same volume into. This is good.
  3. Increase PEEP to 10, plateau is now 24. Driving pressure is the same at 14. You recruited no more lung and hence achieved nothing by increasing PEEP except increasing the distending pressure. Bad PEEP! Too far! Scale back down.
Amato's study is out of the Hopkins MICU, where Brower -- lead author of the original ARDSnet trial -- is director. (Also my alma mater.) There were little mango-colored cards there with the protocol to do these trials.

Anyway, it all makes some sense, but it's unproven. There was another analysis a few months ago by Villar et al which didn't really show it to be all that much better than plateau pressure (maybe very slightly). We shall see.
 

VentMonkey

Family Guy
5,729
5,043
113
So the best-known and best-proven approach to "lung-protective" ventilation for ARDS patients is to limit plateau pressures and tidal volumes, which is proven to improve mortality. The idea is that an ARDS lung has reduced compliance, due to both interstitial changes (lung is less stretchy) and airspace disease (there's less actual lung to ventilate, since some of it is filled with crud). Forcing high pressures/volumes into that causes damage, aka volutrauma.

So we measure the volumes and/or pressures we give. Both are really surrogate measures, though, since we don't know how much lung we're actually ventilating and we don't know how far we can stretch those alveolar before injuring them. Thus we're always looking for better metrics.

Driving pressure was a proposed alternative. It was most popularized by Amato 2015, which reanalyzed the original ARDSnet data looking for a different number called "driving pressure." This is essentially plateau pressure minus the PEEP. Since plateau pressure is the pressure in the alveolar at the end of inspiration, and PEEP is the pressure at the start of inspiration, driving pressure essentially reflects the actual pressure/volume change (delta P) from the breath. Arguably this would better reflect lung protection than plateau alone, which includes PEEP even though the "PEEP air" isn't moving.

You would perform a "driving pressure trial" something like this:
  1. Give a test breath, check driving pressure. Let's say PEEP is 5, plateau 20, so driving pressure is 15.
  2. Increase PEEP a little, check driving pressure after a little while. Let's say PEEP is 7, plateau is now 21. Driving pressure is now LESS at 14. What does this mean? It means you recruited more lung by increasing PEEP. There's more space to put the same volume into. This is good.
  3. Increase PEEP to 10, plateau is now 24. Driving pressure is the same at 14. You recruited no more lung and hence achieved nothing by increasing PEEP except increasing the distending pressure. Bad PEEP! Too far! Scale back down.
Amato's study is out of the Hopkins MICU, where Brower -- lead author of the original ARDSnet trial -- is director. (Also my alma mater.) There were little mango-colored cards there with the protocol to do these trials.

Anyway, it all makes some sense, but it's unproven. There was another analysis a few months ago by Villar et al which didn't really show it to be all that much better than plateau pressure (maybe very slightly). We shall see.
Thanks for that.
 

EpiEMS

Forum Deputy Chief
3,815
1,143
113
My only caveat is that I may "steal" some selected questions to answer on my website. (I'll leave the asker anonymous if desired.)

Brandon, is this a new website, apart from EMS Basics?
 

VentMonkey

Family Guy
5,729
5,043
113
@Brandon O last ARDS question, then hopefully others will have some more of their own as well:

What's your view on proning these patients? Is it something your ICU actively does? I do know some intensivists are more for it, while others question its use in this specific patient population. Are you aware of any recent studies for, or against, proning ARDS patients?

...ok, so I lied; that was three questions.
 

DesertMedic66

Forum Troll
11,268
3,450
113
@Brandon O last ARDS question, then hopefully others will have some more of their own as well:

What's your view on proning these patients? Is it something your ICU actively does? I do know some intensivists are more for it, while others question its use in this specific patient population. Are you aware of any recent studies for, or against, proning ARDS patients?

...ok, so I lied; that was three questions.
My girlfriend works in the ICU and they just recently had to prone an ARDS patient. According to her they have never done it before, they don’t have policies for it, or the equipment for it. They ended up having to rent the bed and have the rep show them how to do it. Per her the patient went from having an SpO2 of 76% on high FiO2 and after about an hour of being prone the patient was 95% on .40 FiO2.
 
OP
OP
Brandon O

Brandon O

Puzzled by facies
1,718
337
83
@Brandon O last ARDS question, then hopefully others will have some more of their own as well:

What's your view on proning these patients? Is it something your ICU actively does? I do know some intensivists are more for it, while others question its use in this specific patient population. Are you aware of any recent studies for, or against, proning ARDS patients?

...ok, so I lied; that was three questions.

Timely! I just published on this.

There are very few interventions proven to improve outcomes in ARDS. Lung-protective (low tidal volume) ventilation is the best demonstrated. In the right subset, there is some weak evidence for paralysis and even weaker evidence for steroids.

Proning has been done for decades now, but the evidence was really on and off. This was until only a few years ago, when the PROSEVA trial finally managed to show a clear reduction in mortality. The trick seems to be: do it early (within the first few days), do it longer (16+ hours a day), and do it in the right population (severe ARDS).

The physiology is interesting and we can talk more about it, but suffice to say that it generally improves oxygenation. Logistically it can be challenging to do safely. The rotating beds that @DesertMedic66 mentioned are pretty sweet, but very expensive and many institutions don't have them. You can prone on a regular bed. The biggest difference is whether you're in a place that does it enough that they're comfortable with it and all the attendant details (nursing a prone patinet, etc), versus somewhere where it's like a Hail Mary and odds are someone's gonna accidentally yank out the central line.

My medium-sized center does and has proned people, but is more in the latter group -- that is, it's something I'd bite my nails about. My last place did it more, particularly in the MICUs. Again, the more you do it, the more comfortable you'll be doing it more. If that makes sense.
 

TXmed

Forum Captain
308
132
43
@Brandon O

How would you manage severe pulmonary hypertension or massive PE on a ventilator ?

Inunderstand obviously PA dilators but more from my point of view which is transport and initial management until definitive care.
 

Aprz

The New Beach Medic
3,029
664
113
I feel silly for asking this one because I feel like I should know it considering I like ECGs a lot. I honestly don't understand how calcium chloride/gluconate works in treating hyperkalemia. I've read the ncbi article on it, but just don't get it. I totally (think) I get that sodium and potassium move back and forth into and out of the cell to cause depolarization. I don't exactly understand calcium role in the depolarization-repolarization cycle.
 

VFlutter

Flight Nurse
3,728
1,264
113
Neuro-critical care. The broad array of intracranial hemorrhages can be hard to differentiate sometimes I.e intracranial, intracerbral, intraparenchymal, etc. Any tips to remember and clarify them? Specifically which ones require more aggressive pressure manamgment like SAH.
 
OP
OP
Brandon O

Brandon O

Puzzled by facies
1,718
337
83
@Brandon O

How would you manage severe pulmonary hypertension or massive PE on a ventilator ?

Inunderstand obviously PA dilators but more from my point of view which is transport and initial management until definitive care.

Splendid question. This is a big topic, and actually a somewhat specialized one, but I'll say what I can.

Pulmonary hypertension is a bit of a bogeyman. Sometimes we seem to ignore it, sometimes we get scared by its very mention. Truly understanding it is not easy, and you have to start by breaking it down by etiology.

You don't need to memorize the names of the WHO groups, but you need to understand how they differ. In brief,

Group 1: Primary/idiopathic. This is Pulmonary Arterial Hypertension, usually due to a not-well-understood degenerative condition that occurs in the unlucky. These people are often on chronic pulmonary vasodilators, in some cases via continuous drip by pump. Like a patient with an LVAD, they need this to live on a minute-by-minute basis (so don't trip over it). They need their vasodilators and should be managed by specialists.

Group 2: Cardiac cause. Pretty easy to understand, this is typical left heart failure stuff.

Group 3: Pulmonary cause. Your COPD or sleep apnea patients.

Group 4: Thrombotic cause. Your PEs.

Group 5: Other/mixed/miscellaneous junk.

You probably recognize that there are disease-specific treatments for many of these. For instance, one of the goals of therapy with lytics or clot retrieval in PE is to reduce the incidence of chronic PH (group 3, see?). But there are some general principles that can be useful:

Right heart strain leading to failure is what you fear. PH causes a lot of effects ranging from vascular remodeling to thrombogenesis, but the acute worry is that the right heart will suddenly poop out. Remember the RV is a low pressure system. Unlike the burly LV, it is a thin-walled chamber accustomed to creating gentle forward pressures only. Normal pulmonary vascular pressures are below 20 or 30 mmHg, compared to the systemic pressures (>100) you know so well. The chunky LV not only has the hypertrophy to squeeze that hard, it also has reserves to go even harder; if systemic BP increases (or there are other obstacles creating afterload, such as outflow obstructions like aortic stenosis), it can push past that, too. Like Jordan at the buzzer, it's got grit.

Unlike Jordan, the RV is more likely to choke in the clutch. If you increase pulmonary arterial pressures sufficiently, it won't squeeze harder; it will simply fail. Cardiac output will drop, the RV will dilate further, and the cycle will worsen. The RV is also easily affected by malperfusion, and cardiogenic shock will hence tend to cause ischemia here, decreasing output and further feeding the cycle.

You may be thinking about right-sided STEMI here, and how we teach to flush with fluids, because you need to help preload the left heart. Implicit here is the idea that you don't really "need" the RV. But this is a very different situation, because in isolated right-sided STEMI, pulmonary vascular pressures are low, and you can passively push fluid through by increasing venous pressures. But in PH the pulmonary vascular pressures are high. You need the RV to do its job in order to push blood against that resistance. If your pulmonary diastolic pressure is 40, and your CVP is 10, I guarantee that blood will not move forward unless the RV can help.

Now, like most things, chronic things are tolerated better than acute things. Over time, the RV can hypertrophy to better withstand the higher pressures. Thus, the COPDer may do better with their PH than the PE patient.

So, to boil this down to basic management:
  1. Follow the Goldilocks approach. These patients cannot withstand "permissive" anything. Hypoxia or hypercarbia will increase pulmonary vascular resistance (by pulmonary vasoconstriction); do not allow it. Hypotension will cause RV ischemia, but hypertension will increase afterload, so avoid both. Use cardiostable drugs, don't get laissez-faire, and keep these people in the box.
  2. Do not fluid overload. The Starling curve of the left heart flattens out eventually, yielding little increase in cardiac output for additional preload. But the Starling curve of the right heart peaks and then actually falls; it is U-shaped, and with overdilation you can suddenly cause the RV to suddenly drop its output. A little fluid is okay in most patients; stop after that.
  3. Beware perturbations in thoracic impedence. Too much PEEP on the ventilator will only add to the external squeeze upon the pulmonary vascular vessels. In fact, the mere switch from negative to positive pressure during intubation can easily cause arrest. The anesthesia gang (who can comment further if they wish) fears these people, and high-risk patients should probably not be going to the OR in a place without rescue bypass available. I once saw a routine post-op patient with undiagnosed PH get a bit hypercarbic in the PACU and suddenly code.
  4. Pressors as needed to maintain perfusion. Vasopressin may be good as it doesn't squeeze the pulm vessels. Norepi is fine. Milrinone may be your best inotrope, in theory. Multiple agents are often needed. Invasive monitoring is often appropriate, and PA catheters can have a role. Pulmonary vasodilators can have a role, although effects probably vary by the PH type.
Helpful?
 
OP
OP
Brandon O

Brandon O

Puzzled by facies
1,718
337
83
I feel silly for asking this one because I feel like I should know it considering I like ECGs a lot. I honestly don't understand how calcium chloride/gluconate works in treating hyperkalemia. I've read the ncbi article on it, but just don't get it. I totally (think) I get that sodium and potassium move back and forth into and out of the cell to cause depolarization. I don't exactly understand calcium role in the depolarization-repolarization cycle.

Good question. As someone said recently on Twitter, there's a fine line between physiology and hand-waving. I bet 99% of people you'll ask won't have a clue how this works. I had to do a lot of digging and ask a lot of people to find an explanation that got even a level or two deeper than the usual, although if you squint hard enough you'll still probably see a hand-wavy firmament.

Here's what I've got. As you know, calcium neither clears potassium nor shifts it, it merely helps stabilize cardiac depolarization in its presence. How?

Cardiac depolarization relies upon the membrane potential. Importantly, this is not the total difference in charged particles intra- vs extra-cellular; it is the local difference in the vicinity of the cell membrane only. Usually it's maintained by the constant outward leak of intracellular potassium, causing an ongoing relative loss of intracellular cations (compared to intracellular negatively-charged proteins, which stay put). This creates the difference in charge across the membrane, which is only a very local effect. The Na-K exchanger contributes just a little bit (by pushing in more K than the Na it pumps out), and Na influx has only a minimal effect, because the membrane isn't very permeable to sodium (compared to high K permeability).

Now, if you have intravascular hyperkalemia, you have more potassium outside the cells, so there's less of a chemical gradient for K efflux, so there's less K movement, so there's less resting membrane potential. So maybe the potential (the gradient) drops from -90 to -80. That's closer to the threshold potential of -75, so it's now easier to depolarize, and you get more irritability.

Giving calcium intravascularly (extracellularly) makes Ca+ bind around the voltage sensor and the pore of all the voltage-gated channels, including the Ca and Na channels (which are responsible for depolarization of pacemaker and ventricular cells, respectively). When it does this, it increases their activation threshold, so they won't open (depolarize) until you reach a higher (less negative) voltage. It desensitizes them. The activation potential therefore moves away from the abnormally-increased resting potential, giving you more breathing room between the two, and hence less excitability. The binding mechanism is not really well understood, but that seems to be the current theory, and I suppose it makes sense, since cations binding around the voltage sensors seem like they'd make it "think" there's a higher (more positive) extracellular voltage, and you'd therefore need a higher (less negative) intracellular voltage to hit the same difference.

Make sense?
 
OP
OP
Brandon O

Brandon O

Puzzled by facies
1,718
337
83
Neuro-critical care. The broad array of intracranial hemorrhages can be hard to differentiate sometimes I.e intracranial, intracerbral, intraparenchymal, etc. Any tips to remember and clarify them? Specifically which ones require more aggressive pressure manamgment like SAH.

There's a lot that can be said, so let me try and make sure I understand your question. Basically, "All these head bleeds -- at the end of the day, what are the distinctions that should change my care in a day to day sense?"

I'll skip the standard stuff you probably understand. When it comes down to it, the majority of the time it works out like this:
  1. You bleed into your head, usually due to trauma.
  2. If it's near the surface (subdural or epidural), either it's big enough to be worth draining, or not. This is often informed by the presence of significant mass effect (midline shift) and clinical findings (obtundation). Often it's not that big.
  3. If not, we mostly stare at them. Frequent neuro checks to look for deterioration (in which case, go back to #2), repeat CT scans (same business). Otherwise just optimization -- avoiding hypotension, hypoglycemia, hypoxia, etc. If they are in bad shape, mostly meaning they are obtunded and therefore we will have a hard time appreciating clinical deterioration (they're already out), or we suspect significantly elevated ICP, maybe we place a bolt or intraventricular drain.
  4. Subarachnoids are the same story, except they have a somewhat higher risk for delayed complications like vasospasm. These are also more likely to be non-traumatic, i.e. aneurysmal rupture, in which case clipping/coiling is usually needed.
  5. Most significant bleeds warrant a short period of seizure prophylaxis. Many minor ones don't. Few need it longer than a week.
  6. Intraparenchymal hemorrhage (aka intracerebral hemorrhage) is the same business, except they are not really drainable. (There is some cutting edge stuff being done with draining very large ones, but you have to go digging in the brain.) These don't do great.
  7. Diffuse axonal injury (DAI) is the same business, but nothing is drainable, just diffuse injury. These don't do great.
Hopefully that helps with categorization -- for most of your care, the similarities are greater than the differences.

BP goals: these are mostly made up. Do what your neurosurgeon wants so you don't get into fights over things you each made up differently. My made-up numbers, pretty consistent with typical recommendations, are:
  • Typical traumatic subdural, epidural, subarachnoid: SBP <160
  • Uncontrolled bleed, i.e. aneurysmal subarachnoid prior to control: SBP <140
  • Subarachnoid with vasospasm: this is its own whole crazy thing and nobody can agree. Usually higher pressures here, plus things like nimodipine.
  • If ICP monitor in place: Usually keep CPP >60, maybe >70-80. With an ICP of, say, 20, this means a MAP >80-100, not a bad goal if you don't have a monitor. Usually NSG will tell you if they want to push the MAP empirically.
  • Avoid ANY hypotension, however. Even short periods you ordinarily might accept are probably not okay. If it's hard to safely hit your sweet spot, drips like nicardipine are your friend.
 

CANMAN

Forum Asst. Chief
805
425
63
Splendid question. This is a big topic, and actually a somewhat specialized one, but I'll say what I can.

Pulmonary hypertension is a bit of a bogeyman. Sometimes we seem to ignore it, sometimes we get scared by its very mention. Truly understanding it is not easy, and you have to start by breaking it down by etiology.

You don't need to memorize the names of the WHO groups, but you need to understand how they differ. In brief,

Group 1: Primary/idiopathic. This is Pulmonary Arterial Hypertension, usually due to a not-well-understood degenerative condition that occurs in the unlucky. These people are often on chronic pulmonary vasodilators, in some cases via continuous drip by pump. Like a patient with an LVAD, they need this to live on a minute-by-minute basis (so don't trip over it). They need their vasodilators and should be managed by specialists.

Group 2: Cardiac cause. Pretty easy to understand, this is typical left heart failure stuff.

Group 3: Pulmonary cause. Your COPD or sleep apnea patients.

Group 4: Thrombotic cause. Your PEs.

Group 5: Other/mixed/miscellaneous junk.

You probably recognize that there are disease-specific treatments for many of these. For instance, one of the goals of therapy with lytics or clot retrieval in PE is to reduce the incidence of chronic PH (group 3, see?). But there are some general principles that can be useful:

Right heart strain leading to failure is what you fear. PH causes a lot of effects ranging from vascular remodeling to thrombogenesis, but the acute worry is that the right heart will suddenly poop out. Remember the RV is a low pressure system. Unlike the burly LV, it is a thin-walled chamber accustomed to creating gentle forward pressures only. Normal pulmonary vascular pressures are below 20 or 30 mmHg, compared to the systemic pressures (>100) you know so well. The chunky LV not only has the hypertrophy to squeeze that hard, it also has reserves to go even harder; if systemic BP increases (or there are other obstacles creating afterload, such as outflow obstructions like aortic stenosis), it can push past that, too. Like Jordan at the buzzer, it's got grit.

Unlike Jordan, the RV is more likely to choke in the clutch. If you increase pulmonary arterial pressures sufficiently, it won't squeeze harder; it will simply fail. Cardiac output will drop, the RV will dilate further, and the cycle will worsen. The RV is also easily affected by malperfusion, and cardiogenic shock will hence tend to cause ischemia here, decreasing output and further feeding the cycle.

You may be thinking about right-sided STEMI here, and how we teach to flush with fluids, because you need to help preload the left heart. Implicit here is the idea that you don't really "need" the RV. But this is a very different situation, because in isolated right-sided STEMI, pulmonary vascular pressures are low, and you can passively push fluid through by increasing venous pressures. But in PH the pulmonary vascular pressures are high. You need the RV to do its job in order to push blood against that resistance. If your pulmonary diastolic pressure is 40, and your CVP is 10, I guarantee that blood will not move forward unless the RV can help.

Now, like most things, chronic things are tolerated better than acute things. Over time, the RV can hypertrophy to better withstand the higher pressures. Thus, the COPDer may do better with their PH than the PE patient.

So, to boil this down to basic management:
  1. Follow the Goldilocks approach. These patients cannot withstand "permissive" anything. Hypoxia or hypercarbia will increase pulmonary vascular resistance (by pulmonary vasoconstriction); do not allow it. Hypotension will cause RV ischemia, but hypertension will increase afterload, so avoid both. Use cardiostable drugs, don't get laissez-faire, and keep these people in the box.
  2. Do not fluid overload. The Starling curve of the left heart flattens out eventually, yielding little increase in cardiac output for additional preload. But the Starling curve of the right heart peaks and then actually falls; it is U-shaped, and with overdilation you can suddenly cause the RV to suddenly drop its output. A little fluid is okay in most patients; stop after that.
  3. Beware perturbations in thoracic impedence. Too much PEEP on the ventilator will only add to the external squeeze upon the pulmonary vascular vessels. In fact, the mere switch from negative to positive pressure during intubation can easily cause arrest. The anesthesia gang (who can comment further if they wish) fears these people, and high-risk patients should probably not be going to the OR in a place without rescue bypass available. I once saw a routine post-op patient with undiagnosed PH get a bit hypercarbic in the PACU and suddenly code.
  4. Pressors as needed to maintain perfusion. Vasopressin may be good as it doesn't squeeze the pulm vessels. Norepi is fine. Milrinone may be your best inotrope, in theory. Multiple agents are often needed. Invasive monitoring is often appropriate, and PA catheters can have a role. Pulmonary vasodilators can have a role, although effects probably vary by the PH type.
Helpful?

Bravo! Thanks for the info!
 
Top