Bicarbonate in Shock: Why Fixing the pH Isn’t Enough
A critical care discussion of why sodium bicarbonate often looks helpful on the blood gas but fails to improve outcomes in vasopressor-dependent metabolic acidosis. The episode breaks down the SODa BIC trial, the surrogate endpoint trap, and when bicarbonate may still have a role in select ICU patients.
Chapter 1
Bicarbonate in Shock: The Reflex to Fix the Numbers
Dr. James Whitfield
It is 3:00 a.m. in the intensive care unit. You are standing in front of a monitor, staring at an arterial blood gas printout. The pH is 7.21, the base excess is minus 6 millimoles per liter, and your patient on high dose norepinephrine is still hypotensive. Your hand almost automatically reaches to order a drip of sodium bicarbonate.
Dr. Elena Rodriguez
It is the classic clinical reflex, James. I mean, every doctor in critical care has felt that exact urge. You see a low pH on the monitor, you know acidemia blunt vascular responsiveness, so you give base to fix the number.
Dr. James Whitfield
Right. I, I cannot tell you how many times on late night call shifts hanging that bicarb infusion felt like the only thing standing between us and total cardiovascular collapse. Because the physiology makes so much sense on paper. Severe metabolic acidosis, defined in these patients as a pH below 7.30 and a base excess less than or equal to minus 4, it genuinely depresses myocardial contractility. It stops vasopressors from working effectively at the receptor level.
Dr. Elena Rodriguez
But biologically, what are we actually doing at the cellular level when we push that bicarbonate? On paper, buffering extracellular hydrogen ions should restore catecholamine sensitivity. But bicarb, when it combines with excess hydrogen ions, generates carbon dioxide and water. And if the lungs cannot blow off that newly generated carbon dioxide fast enough, that CO2 diffuses right back across the cell membrane into the intracellular space.
Dr. James Whitfield
So you end up with paradoxical intracellular acidosis, even while your blood gas printout tells you the blood pH looks better.
Dr. Elena Rodriguez
Exactly. You fix the surrogate number in the blood tube, but the cells themselves might actually be struggling worse than before. And that tension, that exact gap between fixing a lab value and helping a patient, is why researchers set up the SODa BIC trial.
Dr. James Whitfield
And SODa BIC was a massive undertaking. A pragmatic, adaptive, double blind randomized trial across 55 ICUs in 7 countries. They enrolled 500 adults with metabolic acidosis who were actively on vasopressors. One group got a targeted 5 hour sodium bicarbonate infusion titrated to get that pH back up to at least 7.30, while the control group got a matching 5 percent dextrose placebo drip.
Dr. Elena Rodriguez
Double blind, fully pragmatic, testing the exact intervention thousands of clinicians deploy every single day. So what actually happened when the numbers came in?
Chapter 2
The SODa BIC Trial: Lab Surrogates vs Hard Clinical Outcomes
Dr. James Whitfield
Well, the primary outcome was Major Adverse Kidney Events at 30 days, or MAKE30. And when they looked at the primary endpoint, 40.2% of patients in the sodium bicarbonate group hit a major adverse kidney event, compared to 39.4% in the placebo group. That is 98 out of 244 versus 100 out of 254.
Dr. Elena Rodriguez
So an adjusted difference of 1.2 percentage points, with a 95% confidence interval running from minus 7.1 to 9.4, and a P value of 0.78. That is essentially identical.
Dr. James Whitfield
Completely flat. Absolutely zero statistically significant difference. And it did not stop there. When you look at secondary endpoints, renal replacement therapy was required in 16.8% of the bicarbonate group compared to 20.9% in the placebo group, with an adjusted difference of minus 3.9 percentage points, 95% confidence interval minus 10.6 to 2.7. And 30 day in hospital mortality was 25.4% with bicarb versus 24.0% with placebo, an adjusted difference of 1.8 percentage points, 95% confidence interval minus 5.6 to 9.2.
Dr. Elena Rodriguez
And safety signals? Did the bicarbonate group show any signs of harm?
Dr. James Whitfield
Adverse events were reported in 4 patients in the bicarbonate arm, which is 1.6%, versus 0 in the placebo group, with a P value of 0.06. So while overt harm was rare, there was certainly no sign of clinical benefit.
Dr. Elena Rodriguez
This is such a classic example of the surrogate endpoint trap in critical care medicine. We look at the arterial blood gas, we see a pH climbing back over 7.30, and our brains tell us we are fixing the problem. But normalizing a serum biomarker on a lab slip is simply not the same thing as altering underlying disease biology.
Dr. James Whitfield
It really forces us to re examine our bedside habits. Routine, protocolized bicarbonate infusions for unselected patients in septic shock or vasopressor dependent metabolic acidosis just do not hold up under rigorous randomized trial conditions.
Dr. Elena Rodriguez
Now, to be clear for clinicians listening, this does not mean bicarb is dead everywhere in the ICU. We still have data supporting its targeted use in severe acute kidney injury with profound acidemia where pH drops below 7.20, or in specific toxicological overdoses like tricyclic antidepressants.
Dr. James Whitfield
Right. But for the general patient in shock, SODa BIC tells us to stop treating the printout and focus on treating the root cause of the hypoperfusion instead.
Dr. Elena Rodriguez
A great lesson in cellular reality over lab surrogates. Good chatting with you, James.
Dr. James Whitfield
Talk soon, Elena.