Probably a simple question about sepsis...

[lightthelamp4]

So I think I may be over thinking this or missing something very basic. I know that an SvO2 > 70% is one of the treatment targets in sepsis. What I dont get is why in sepsis would SvO2 drop to begin with if you have a distributive shock and you cant perfuse tissue adequately? I would think that if you have less peripheral perfusion that you would have a higher SvO2
I (believe) I understand that SvO2 will decrease when you have higher metabolic O2 demand in the tissues, and that when SvO2 drops to a certain point the metabolic demands of the tissue will force it into anaerobic metabolism; but if you cant deliver oxygen (which is my conception of distributive shock) how would you lower SvO2 by increased O2 consumption?

---

Members don't see this ad.

 

[thehundredthone]

Tissue perfusion does fall but not to zero. That would be ischemia (which technically can happen in the worst case). In any case, what happens is that tissues extract more oxygen from the blood around them given the relative lack/sluggishness of blood flow and so the SvO2 drops. This is after the initial compensation by the system, which is to increase CO. The tissues don't care about the reason for hypoxia, they pull out whatever they can. If this still doesn't do enough, they begin to switch over to anaerobic metabolism (and then lactate begins to reflect this). ScvO2 is used in practice (from the central line) to monitor the effectiveness of resuscitation in sepsis where you're trying to overcome the systemic vasodilation by expanding the plasma volume through large boluses, and by vasopressors when the former is inadequate.

However, like you said, SvO2 can be high in sepsis too, due to hyperdynamic circulation (early sepsis) or microvascular shunting (despite end organ hypoxia/damage), or sometimes when your ventilator FiO2 is high, in which case it's not as useful a marker/goal. That's why we rely on more than just one parameter and monitor for signs of possible end organ damage through ABGs, renal and liver function tests, arterial BP monitoring, etc.

 

[lightthelamp4]

awesome, thanks so much !

 

[mrpankration]

A minor point, but if you want to be precise and impress your ICU attendings, is that a Sv02 is actually drawn from the pulmonary artery from a PA catheter. The vast majority of patients with severe sepsis/septic shock will not have PA catheters but will have CVCs which typically sit right near the cavoatrial junction. The measurement that is taken from the CVC is usually from the SVC and to be precise is a Scv02. The only real point for distinction is that Scv02 may be slightly higher than the Sv02 but is usually close enough to where is acceptable to use as a surrogate.

Another point to make about Scv02 in severe sepsis/septic shock is that the value is also affected by cell dysoxia, where there is actually an intracellular impairment in oxygen utilization, this usually happens at the mitochondrial level. So even if you are perfusing tissue, the cells have essentially been prevented from being able to utilize the oxygen that is being delivered to them. That can be a reason for high Scv02 in sepsis.

The last point, and perhaps the most important, is that routine catheter-based measurements such as CVP and Scv02 are falling out of favor based on the results of the ProCESS trial that came out a few months ago, which showed that these measurements as carried out as part of a EGDT bundle did not improve mortality (as they had shown to do previously in the 2001 Rivers' EGDT trial).

Hope that helps a little!

 

[Instatewaiter]

SWAN EVERYBODY!

 

[Bostonredsox]

SWAN EVERYBODY!

I totally would. Too bad echo is so good now that everyone frowns on me sticking additional plastic into pts..

 

[jdh71]

So I think I may be over thinking this or missing something very basic. I know that an SvO2 > 70% is one of the treatment targets in sepsis. What I dont get is why in sepsis would SvO2 drop to begin with if you have a distributive shock and you cant perfuse tissue adequately? I would think that if you have less peripheral perfusion that you would have a higher SvO2
I (believe) I understand that SvO2 will decrease when you have higher metabolic O2 demand in the tissues, and that when SvO2 drops to a certain point the metabolic demands of the tissue will force it into anaerobic metabolism; but if you cant deliver oxygen (which is my conception of distributive shock) how would you lower SvO2 by increased O2 consumption?

This isn't a tough or complicated concept. Until tissue dies and/or is poisoned, it will increase it's demand for oxygen when sick. That's it. It drops because the tissues are pulling more oxygen.

Though Sv02 isn't a good marker for end resuscitation UNLESS you're trying to sell someone a central venous catheter.

If the lactate is clearing, then who cares if you're Svo2 is still below normal. The tissues are getting what they need. It's not broke and doesn't need to be fixed.

 

[jdh71]

I totally would.

I wouldn't. There is too much data that says putting a swan in these people does exactly DICK to make them better (one study even showed worse outcomes with a swan - though it kind of boggles my mind how it could make someone worse). All a swan does is allow you to THINK you are doing something good by treating numbers - at best a single snap shot of the picture at any given time. It's would like like trying to fight a war by thinking if you take a picture every couple of hours of the battlefield it will help you fight the war better. If you're a good general you should have a good idea of what is going on with the battlefield and if you need the photograph to tell you, I dare say, you aren't paying enough attention.

Physicians treat PATIENTS not numbers.

Swans belong in sick right hearts with something else going on so you can know what's happening with the pulmonary pressures with any given intervention on a minute to minute basis. It's just a fancy arterial line, and that's its only real utility.

 

[Instatewaiter]

I wouldn't. There is too much data that says putting a swan in these people does exactly DICK to make them better (one study even showed worse outcomes with a swan - though it kind of boggles my mind how it could make someone worse). All a swan does is allow you to THINK you are doing something good by treating numbers - at best a single snap shot of the picture at any given time. It's would like like trying to fight a war by thinking if you take a picture every couple of hours of the battlefield it will help you fight the war better. If you're a good general you should have a good idea of what is going on with the battlefield and if you need the photograph to tell you, I dare say, you aren't paying enough attention.

Physicians treat PATIENTS not numbers.

Swans belong in sick right hearts with something else going on so you can know what's happening with the pulmonary pressures with any given intervention on a minute to minute basis. It's just a fancy arterial line, and that's its only real utility.

In sepsis, sure. Mixed picture or cardiogenic shock, I completely disagree. Having a PA catheter allows you to move much more quickly. The issue is that most people don't really know what to do with the numbers.

Having come from a place that never used PA catheters to a place that almost always uses them, I have found the utility of a PA catheter and the speed at which you catch and prevent people from going on the downward spiral.

 

[jdh71]

In sepsis, sure. Mixed picture or cardiogenic shock, I completely disagree. Having a PA catheter allows you to move much more quickly. The issue is that most people don't really know what to do with the numbers.

Having come from a place that never used PA catheters to a place that almost always uses them, I have found the utility of a PA catheter and the speed at which you catch and prevent people from going on the downward spiral.

Wasn't sepsis the context of this discussion? Though even in a mixed picture, unless the heart failure is actually pretty bad (EF's 20 or less) you can treat the lower function heart about the same for all practical purposes. If the lactate isn't clearing and the LV function looks like **** on a bedside U/S add an ionotrope. If you're still having trouble, probably time for a swan. But putting them in everyone to find that case or two a year like that isn't going to be worth it overall.

If my cardiologists want to use them in cardiogenic shock patients that is their business.

 

[Bostonredsox]

I wouldn't. There is too much data that says putting a swan in these people does exactly DICK to make them better (one study even showed worse outcomes with a swan - though it kind of boggles my mind how it could make someone worse). All a swan does is allow you to THINK you are doing something good by treating numbers - at best a single snap shot of the picture at any given time. It's would like like trying to fight a war by thinking ate if you take a picture every couple of hours of the battlefield it will help you fight the war better. If you're a good general you should have a good idea of what is going on with the battlefield and if you need the photograph to tell you, I dare say, you aren't paying enough attention.

Physicians treat PATIENTS not numbers.

Swans belong in sick right hearts with something else going on so you can know what's happening with the pulmonary pressures with any given intervention on a minute to minute basis. It's just a fancy arterial line, and that's its only real utility.

I wasn't looking to enter into a debate on the utility of swans, which we have previously discussed ad nauseum. I just sarcastically said I would put them in everyone if there was no echo available. Because I enjoy doing procedures and looking at real time data compared with a bland, stick a probe on the IVC and take a 2 sec picture. But as the data shows the swan doesn't help much, certainly in sepsis, and it does have associated complications although exceedingly rare today, I don't. Hence the sarcasm.

But I agree with instate. Our bypass pts have them and I use them in mixed picture cardiogenic shocks with bad cor pulmonale. I don't use them in septic shock as they don't provide me with any additional actionable information.

 

[souljah1]

ScvO2 is low in septic shock most often on presentation due to concomitant hypovolemia. You can prove this to yourself by fluid bolusing these patients and checking a ScvO2 before and after. Watch the fluid bolus improve your C.I. ScvO2 can also be low in septic shock after adequate fluid resuscitation because your C.I. is depressed due to a sepsis-induced cardiomyopathy.

 

[jdh71]

I wasn't looking to enter into a debate on the utility of swans, which we have previously discussed ad nauseum. I just sarcastically said I would put them in everyone if there was no echo available. Because I enjoy doing procedures and looking at real time data compared with a bland, stick a probe on the IVC and take a 2 sec picture. But as the data shows the swan doesn't help much, certainly in sepsis, and it does have associated complications although exceedingly rare today, I don't. Hence the sarcasm.

But I agree with instate. Our bypass pts have them and I use them in mixed picture cardiogenic shocks with bad cor pulmonale. I don't use them in septic shock as they don't provide me with any additional actionable information.

Your bypass patients have them in because anesthesia put them in, in the OR, and I can't speak to their utility in that setting because I'm not an anesthesiologist, but I assume, in that kind of acute setting, they are better for patients than not having them. If they come out of the OR with them, no reason not to use them, but they are not put in for post-op care, but rather intra-op care. I'd bet that outcomes wouldn't be statistically different between a group where you pulled at the end of the case and those where you did not, but you'll never find anyone brave enough to try and probably no IRB willing to let something like that happen, plus I can't imagine that kind of informed consent discussion . . .

 

[sluggs]

Paul Marik had a nice editorial/review/opinion piece in Chest in June questioning the utility of using scvo2 as a resuscitation benchmark and advocated using a cardiac output target of 2.5. He also specified that one ought use non-invasive [echo] or minimally invasive [?picco, flo track] to assess. Physiologically, to me, it makes sense that a good majority (can't give you a better number) of patients who are adequately resuscitated with fluids and pressers and still have a low scvo2 would have low CO and benefit from inotropes, However, there are also patients with scvo2 around 70 who still have poor CO and need inotropes.
Also, I think it is important to think about the JAMA study that purported that lactate clearance is equivalent to scvo2 normalization as a resuscitation benchmark: first, only 10% of the patients in either arm actually had a low scvo2 that would have "required" inotropy, and this was not statistically significant (i.e. in patients who responded to fluids and inotropes and were improving, lactate clearance was just as good a marker as scvo2 that the resuscitation was going well, but this may not be the case in patients who are doing poorly despite fluid and pressor resusc).
Finally, think about the patient who initially clears some lactate (a mere 10% was the benchmark in the JAMA paper), but then is under-fluid resuscitated and over-pressored (this does happen, especially at "outside hospital"), and starts to go south (often late gut ischemia). Cardiac output might help pick this up sooner and more sensitively than scvo2.
Much of this is opinion, but in light of PROCESS, I think that normalizing scvo2 (especially with blood, too) is no longer an evidence-validated strategy, and we need better surrogates for capturing patients in whom a "septic heart" actually mandates resuscitation.

 

[nlax30]

I found it annoying when our hospital implemented our "goal directed therapy" and to meet the pure number end points I'd routinely come in to find a septic shock patient on dobutamine because their Scv02 was like 63.....

 

[sluggs]

I had a patient a few weeks ago who was a bit septic, but who was in shock from acute on chronic heart failure (or decompensated heart failure) with an EF of 15%. Her ScVO2 was 64!
Bad marker!