Sodium Bicarbonate for Metabolic Acidosis in the ICU: Results of a Pilot Randomized Double-Blind Clinical Trial.
Serpa Neto A., Fujii T., McNamara M., Moore J., Young PJ., Peake S., Bailey M., Hodgson C., Higgins AM., See EJ., Secombe P., Campbell L., Young M., Maeda M., Pilcher D., Nichol A., Deane A., Licari E., White K., French C., Shehabi Y., Cross A., Maiden M., Kadam U., El Khawas K., Cooper J., Bellomo R., Udy A.
ObjectivesTo identify the best population, design of the intervention, and to assess between-group biochemical separation, in preparation for a future phase III trial.DesignInvestigator-initiated, parallel-group, pilot randomized double-blind trial.SettingEight ICUs in Australia, New Zealand, and Japan, with participants recruited from April 2021 to August 2022.PatientsThirty patients greater than or equal to 18 years, within 48 hours of admission to the ICU, receiving a vasopressor, and with metabolic acidosis (pH < 7.30, base excess [BE] < -4 mEq/L, and Paco2 < 45 mm Hg).InterventionsSodium bicarbonate or placebo (5% dextrose).Measurements and main resultThe primary feasibility aim was to assess eligibility, recruitment rate, protocol compliance, and acid-base group separation. The primary clinical outcome was the number of hours alive and free of vasopressors on day 7. The recruitment rate and the enrollment-to-screening ratio were 1.9 patients per month and 0.13 patients, respectively. Time until BE correction (median difference, -45.86 [95% CI, -63.11 to -28.61] hr; p < 0.001) and pH correction (median difference, -10.69 [95% CI, -19.16 to -2.22] hr; p = 0.020) were shorter in the sodium bicarbonate group, and mean bicarbonate levels in the first 24 hours were higher (median difference, 6.50 [95% CI, 4.18 to 8.82] mmol/L; p < 0.001). Seven days after randomization, patients in the sodium bicarbonate and placebo group had a median of 132.2 (85.6-139.1) and 97.1 (69.3-132.4) hours alive and free of vasopressor, respectively (median difference, 35.07 [95% CI, -9.14 to 79.28]; p = 0.131). Recurrence of metabolic acidosis in the first 7 days of follow-up was lower in the sodium bicarbonate group (3 [20.0%] vs. 15 [100.0%]; p < 0.001). No adverse events were reported.ConclusionsThe findings confirm the feasibility of a larger phase III sodium bicarbonate trial; eligibility criteria may require modification to facilitate recruitment.