BICAR-ICU: How Should this Study Affect Care in the ED?

By Marco Torres September 10, 2018 0 comments

During a busy resuscitation it behooves the ED resuscitationist to avoid ordering therapies that have no clear benefit to their patients. Our nurses are overburdened, and adding interventions ‘just to be safe’ or ‘because we always do it’ is a great way to overwhelm our nursing colleagues. Bicarb administration is a contentious issue and many clinicians consider its use to fall firmly into the no-benefit camp, even when used in the management of severe metabolic acidosis. There have been no studies to date evaluating clinical outcomes with the use of sodium bicarbonate infusion therapy for severe metabolic acidemia, until now. This post is a review of the recently published BICAR-ICU trial

Bicarb Administration in the ED Typically Occurs in one of 2 Settings:

‘Crash cart bicarb’: the 50cc 8.4% tan packaged syringes
‘Normal bicarb’ infusion, with 3 amps of bicarb added to sterile water or D5W, giving a final concentration of 150meq/l NaCO3, with or without the 5% dextrose. This is 1.3% Bicarb.

There is a body of evidence that suggests the rapid administration of 8.4% bicarb for the management of severe metabolic acidosis is not helpful (Resuscitation Sequence Intubation: pH kills). There is also a plausible concern that it could be potentially harmful by causing a paradoxical intracellular acidosis, or by impairing oxygen delivery due to a shift in the oxyhemoglobin dissociation curve. (Post Intubation Hypotension: The Aah Shite mnemonic).

The potential harms of a rapid bicarb push are likely related to the speed of bicarb delivery. Giving the same amount of bicarb over a longer time period can conceivably avoid any paradoxical intracellular acidosis and may not mess with the oxyhemoglobin dissociation curve.

With the swelling numbers of drug shortages in the US, Bicarb is now a limited commodity in some hospital systems. Consuming 3 amps of bicarb to make a 1000ml normal bicarb infusion for the management of severe metabolic acidosis is a common practice. It would be nice to have a robust body of evidence to support this practice. The BICAR-ICU trial is the next best thing:

BICAR-ICU: Sodium bicarbonate therapy for patients with severe metabolic acidaemia in the intensive care unit (BICAR-ICU): a multicenter, open label, randomized controlled, phase 3 trial. Lancet, 2018. PMID: 29910040

Multicenter open label trial in 26 ICU’s in France. Enrollment period May 5 2015 – May 7 2017.
389 patients with severe acidemia (pH<7.2, pCO2 <45mmHg, HCO3 <20) randomized to receive a bicarb infusion to maintain pH >7.3.
Primary Outcome: Composite of 28 day all cause mortality, and at least one organ failure at day 7.
Patients were assigned within 48 hours of ICU admission.
Indications for renal replacement therapy were standardized: At 24h after inclusion RRT occurred for 2/3 criteria: Urine output <0.3ml/kg/h for at least 24h, pH <7.2 despite resuscitation, Potassium >6.5mmol/l. Dialysis includes the use of bicarb as a buffer, so those patients receiving dialysis in either study group received additional bicarb.
Both groups were well matched. 9% of each group were post arrest patients, 51% (control) and 55% (bicarb) had septic shock, and 21% (control) and 23% (bicarb) had hemorrhagic shock.
The majority of patients were on mechanical ventilation (82% control, 84% bicarb) and pressors (80% control, 79% bicarb) at the time of enrollment.
Of 942 patients assessed for eligibility 542 were excluded: These included 41 patients with ketoacidosis, 69 patients with chronic renal failure, 47 patients needing immediate renal replacement therapy, 109 already received bicarb. Additional exclusion criteria included exogenous acid load ( ie ASA OD)
An a priori planned analysis of patients with Acute Kidney Injury Network (AKIN) score of 2 or 3 was performed. (46% (control) and 47% (bicarb))

Acute Kidney Injury Network (AKIN) Score

Creatinine 1.5 – 1.9 times higher than baseline, or bumped by >0.3mg/dl. OR Urine output <0.5ml/kg/h for 6 hours.
Creatinine 2 – 2.9 times higher than baseline OR Urine output <0.5ml/kg/h for 12 hours
Creatinine greater than 3 times baseline or greater than 4 mg/dl with acute rise of >0.5mg/dl. OR urine output <0.3ml/kg/h times 24h or anuric for 12 hours.

The bicarb concentration was different from the normal bicarb infusion concentration in the US.

Bicarb concentration: 4.2%.
The study protocol called for an infusion of 125-250 ml of 4.2% bicarb over 30 min, with no more than 1000ml given in 24h. Titrated to a pH of >7.3
ABG done 1-4h after each infusion.
Standard crash cart bicarb is 8.4% in 50cc.
A typical normal bicarb infusion in the US is 3 amps of 8.4% Bicarb in D5W, or sterile water.
This gives a bicarb concentration of 1.3%

What is the equivalent volume of normal bicarb compared to 4.2% Bicarb?

125-250ml of 4.2% is the bicarb equivalent of 404-808ml of a normal bicarb infusion.
1000ml of 4.2% bicarb is the bicarb equivalent of 3230ml of a normal bicarb infusion.

Results:

Patients were moderately acidotic. Median (and interquartile range) pH values were 7.15 (7.11-7.18) control and 7.15 (7.09-7.18) bicarb.
Target pH reached: 26% control, 60% bicarb
Cumulative bicarb intake from enrollment to 24 hours: 500 ml(interquartile range 250-750) This is equivalent to 1615ml of a normal bicarb 1.3% infusion (IQR 808-2423)
Primary outcome (Death or at least one organ failure at day 7): 71% control, 66% bicarb (p 0.24)

AKIN Score 2 or 3 Patients:

Death or at least one organ failure at day 7: 82% control, 70% bicarb
28 day survival: 46% control vs 63% bicarb (95% CI 52-72 vs 35-55 p 0.0283)
Use of renal replacement therapy during ICU stay: 73% vs 51% (p 0.002)

Significant Secondary Outcomes:

Use of renal replacement therapy during ICU stay: 52% control, 35% bicarb. (p 0.0009)
Time from enrollment to initiation of RRT: 7 vs 19h (p < 0.0001)
There was no difference in duration of vasopressor use, or in the number of vasopressor free days between the 2 groups.

Potential Harms of Intervention:

Bicarb group had higher rates of metabolic alkalosis, hypernatremia, and hypocalcemia without any life threatening complications reported.

Strengths:

Randomized controlled study
Pragmatic intervention (bicarb or no bicarb)

Weaknesses:

Single country study
Bicarb concentration is not commonly used in the US
ICU, not ED based study, with enrollment up to 24h after admission (this raises applicability issue when using this intervention in the ED)
Open label study
Different etiologies for metabolic acidosis
Unknown number of profoundly acidotic patients (6.8-6.9 range)
109 patients were excluded because they had already received bicarb. These may have been sicker patients that could have seen more benefit.

How will I use this information on my next patient with profound metabolic acidosis?

A unique challenge we face in the ED is to make a treatment decision on our critically ill patients, then defend that decision to the admitting intensivist, and then accept recommendations for further interventions from the intensivist before the patient is transferred to the ICU. We need to know the rationale for giving / holding a bicarb infusion, and to know whether it will be a neutral / helpful / harmful intervention.

This study can help us to make slightly more educated guesses on bicarb administration:

Bicarb doesn’t replace the need to correct the cause of the underlying metabolic acidosis. Treat the cause, then consider bicarb. (especially in lactic acidosis caused by hypoperfusion)
In patients with established, or new kidney injury, there is a clear signal toward benefit. We can make an estimate of the AKIN score, however do not have the luxury of knowing the urine output for the preceding 6 / 12 / 24 hours. For those patients with an AKIN score of 2-3, I will give bicarb.
When I give bicarb, I will give normal bicarb (1.3%) 500cc over 30 min. (125-250ml of 4.2% is the equivalent bicarb load of 404-808 ml normal bicarb)
When asked by my intensivist colleagues to give an amp or 2 of 8.4% bicarb for my acidotic patients, I will counter with an offer of 500cc normal bicarb.
I will make my bicarb decision early, and factor in the 500cc of normal bicarb into the initial fluid resuscitation.
I will be aware that my ED population is different than the French ICU population, and that 1.3% normal bicarb may not be an exact substitute for 4.2% bicarb, even when giving the same total amount of bicarb.

References:

BICAR-ICU: Sodium bicarbonate therapy for patients with severe metabolic acidaemia in the intensive care unit (BICAR-ICU): a multicenter, open label, randomized controlled, phase 3 trial. Lancet, 2018. PMID: 29910040