(288) Impact of an IV Bicarbonate Shortage in a Sample of Adult ICUs: Interrupted Time-Series Analysis

Introduction: Limited research exists to guide the use of sodium bicarbonate among adults admitted to ICUs. From June to October 2017, there was a global shortage of I.V. sodium bicarbonate. Institutions like ours enacted conservation measures during this time period. We assessed the effect of the 2017 shortage of I.V. sodium bicarbonate on the use of sodium bicarbonate, and the incidence of severe acidemia (pH ≥7.20), hyperkalemia (serum K ≥6 mmol/L), renal replacement therapy use, and ICU mortality.

Methods: This was an interrupted time series analysis among adults admitted to ICUs in Alberta, Canada between June 1, 2015 and December 31, 2019. The primary data source was eCritical, a population-based electronic health record and data registry for all Alberta ICUs. Individual patient data were aggregated and examined at the ICU-level per month. We included admissions with blood gas pH < 7.3 at any time during their ICU stay. The exposure was a bolus or infusion I.V. bicarbonate prescription. The primary outcome was the proportion of the sample treated with I.V. bicarbonate.

Results: 19,293 admissions met inclusion criteria among 19 ICUs in 15 sites. The median age was 63 years (IQR 52 – 72 years), 36% were female, and median APACHE II was 23 (IQR 18 – 30). Among all admissions, 31% received bicarbonate in the pre-shortage period, 22% during the shortage and 29% post-shortage. Among patients with severe acidemia, 51% received bicarbonate pre-shortage, which fell to 41% during the shortage and was maintained at 41% post-shortage. Among patients with hyperkalemia, 45% received bicarbonate pre-shortage, which fell to 34% during the shortage, and rose to 45% post-shortage. The proportion of patients with severe acidemia was 42% pre-shortage, 42% during shortage, and 40% post-shortage. The proportion of patients with hyperkalemia was 15% pre-shortage, 16% during shortage, and 17% post-shortage. The proportion of patients requiring renal replacement therapy was 16% pre-shortage, 17% during shortage, and 16% post-shortage. There was no difference in ICU mortality.

Conclusions: The I.V. sodium bicarbonate shortage was associated with its reduced administration, including among patients with severe acidemia and hyperkalemia, two conditions where sodium bicarbonate administration is a common practice.