BICAR-ICU -- Will it change your practice?

[lymphocyte]

Multicenter RCT involving 389 patients within 48 hours of ICU admission.

Inclusion required the presence of metabolic acidosis (pH < 7.20, PaCO2 < 45) plus either SOFA score >3 or lactate >2.

Intervention was 4.2% bicarb infusion aiming for pH > 7.3 in 150-250mL aliquots upto 1000mL maximum a day.

Main findings:

1. For a pre-specified subgroup of patients with AKIN II-III renal injury, bicarbonate improved mortality (46% vs. 63%, p=0.017).

2. Bicarbonate reduced the need for dialysis from 52% to 35% (p=0.0009).

I think along with the conclusions of SMART-MED and SMART-SURG, we're getting better at (or at least more mindful about) reno-protective fluid resuscitation.

I'm just curious what other people think and/or what their practice might be.

---

Members don't see this ad.

 

[Hernandez]

Probably won’t change anything substantially over long term, the control group had a very low threshold for rrt. There was no difference in primary end points and most labs were very similar at 48 hours

 

[lymphocyte]

Probably won’t change anything substantially over long term, the control group had a very low threshold for rrt. There was no difference in primary end points and most labs were very similar at 48 hours

The sub-group mortality benefit is very hard to ignore, and the trial was pretty underpowered (hence the wonky composite end-point). No clinically significant harm. So... why not? At least in patients with AKI.

 

[Hernandez]

The sub-group mortality benefit is very hard to ignore, and the trial was pretty underpowered (hence the wonky composite end-point). No clinically significant harm. So... why not? At least in patients with AKI.

and a random sicu study showed mortality benefit for intensive glycemic control. There was a trend toward mortality benefit but not significant. Only real improvement was less RRT. But they don’t talk about long term renal function and the way they had the control group set up they had a really low threshold for rrt which I think skew the data. So I see no harms or benefits at this time, I will not be changing practice patterns until better data is available.

 

[lymphocyte]

and a random sicu study showed mortality benefit for intensive glycemic control.

What does that have to do with anything we're talking about now? It's a red herring. Besides:

1. The Leuven surgical trial was a single-centre trial. This trial was a multi-centre trial with pre-specified sub-groups.

2. NICE-SUGAR and the Leuven medical trial didn't look at patients on parenteral nutrition, unlike the Leuven surgical trial. The real take-away from the Leuven surgical trial (in light of the border evidence) was that people on parenteral nutrition need more intensive glucose control, and that's been borne out by a fair few observational and retrospective trials. I think the Griesdale meta-analysis supports this position as well.

NICE-SUGAR is way too often ignorantly cited for "OMG scepticism"!, when the real sceptics understand the strengths and limitations of the Leuven trials.

So, again, apart from a red herring, you haven't commented on the mortality benefit in patient's with AKI. Several lines of evidence (pathophysiological, observational, and now a prospective multi-centre RCT) point to the fact that bicarb is significantly reno-protective in this patient population.

 

[Hernandez]

Red herring is a way to say I’m not impressed with the data at this time, I need to see another study before I fully alter my practice pattern.

 

[deleted547339]

I’m pretty skeptical.

 

[cbrons]

What does that have to do with anything we're talking about now? It's a red herring. Besides:

1. The Leuven surgical trial was a single-centre trial. This trial was a multi-centre trial with pre-specified sub-groups.

2. NICE-SUGAR and the Leuven medical trial didn't look at patients on parenteral nutrition, unlike the Leuven surgical trial. The real take-away from the Leuven surgical trial (in light of the border evidence) was that people on parenteral nutrition need more intensive glucose control, and that's been borne out by a fair few observational and retrospective trials. I think the Griesdale meta-analysis supports this position as well.

NICE-SUGAR is way too often ignorantly cited for "OMG scepticism"!, when the real sceptics understand the strengths and limitations of the Leuven trials.

So, again, apart from a red herring, you haven't commented on the mortality benefit in patient's with AKI. Several lines of evidence (pathophysiological, observational, and now a prospective multi-centre RCT) point to the fact that bicarb is significantly reno-protective in this patient population.

Bolded and underlined is the only evidence I care about. There could be a million observational studies and meta analyses on said studies and they wouldn't be worth anything.

 

[lymphocyte]

I’m pretty skeptical.

Why?

 

[lymphocyte]

Bolded and underlined is the only evidence I care about. There could be a million observational studies and meta analyses on said studies and they wouldn't be worth anything.

That's a pretty silly position to take, especially in intensive care. For example, may I ask what transfusion triggers you use and why?

Jean-Louis Vincent wrote a lovely and provocative article on this very topic.

Vincent JL. We should abandon randomized controlled trials in the intensive care unit. Crit Care Med. 2010;38(10 Suppl):S534–8.

 

[cbrons]

That's a pretty silly position to take, especially in intensive care. For example, may I ask what transfusion triggers you use and why?

Jean-Louis Vincent wrote a lovely and provocative article on this very topic.

Vincent JL. We should abandon randomized controlled trials in the intensive care unit. Crit Care Med. 2010;38(10 Suppl):S534–8.

that article isn't lovely its stupid

 

[deleted547339]

Why?

The overwhelming majority of data suggests that alkali therapy in anion gap metabolic acidosis is either not helpful or harmful. Primary endpoint in this study is negative, and it’s already a composite endpoint, which significantly weakens the data (in my opinion). If you want to tell me that virgos do better getting bicarbonate, I’d be a little skeptical too. I get that the subgroup was a priori, but it’s still a subgroup analysis.

I’m not saying that in 5 years my practice might change with more data, but a single subgroup showing benefit isn’t going to change my practice.

 

[Vigileo]

Interesting study with a decent design though it's underpowered (15% absolute difference between groups for power calculation?). Limited by the amount of crossover between groups, and even if it's a priori, I agree with TimesNewRoman that it doesn't push me to change given the risk for type I error. Bicarb remains in my arsenal but mostly gets relegated to a hail Mary for refractory acidemia and shock mostly.

 

[jdh71]

These composite end points make me want to kick puppies.

They exist so people making studies that cannot show real statistical significance can find a place for the crap that is being thrown at the wall to stick.

 

[deleted547339]

These composite end points make me want to kick puppies.

They exist so people making studies that cannot show real statistical significance can find a place for the crap that is being thrown at the wall to stick.

So you’re saying you hate cardiologists? Lol.

The thought of a 10,000+ patient study with a composite end point makes my head explode.

 

[jdh71]

So you’re saying you hate cardiologists? Lol.

The thought of a 10,000+ patient study with a composite end point makes my head explode.

Of course I hate cardiologists. But there is no need to start bashing those guys in this thread in the specific.

 

[sluggs]

If you resuscitate with LR or plasmalyte, you rarely have patients who meet the inclusion criteria for the study... they just don't get or stay that acidotic!