No-Change From Baseline Bicarbonate Protects Against Pressors in High-Lactate, Normotensive Sepsis

Introduction: Occult shock is defined as hypoperfusion without hypotension. Differentiating between hyperlactatemia due to occult shock and other causes is a clinical challenge. Lactate correlates poorly with markers of acidosis in sepsis, cardiogenic shock, trauma, surgical ICU patients, and in cord blood. This discrepancy could be useful in differentiating between causes of hyperlactatemia. The primary objective of this study was to evaluate whether no-drop in serum bicarbonate (bicarb) from baseline protects against needing pressors in ED patients with sepsis, lactate ≥4 mmol/L, and initial normotension.


Methods: This is a retrospective cohort study of ED patients ≥18 years in the 21-hospital Kaiser Permanente, Northern California system between 1/1/2012-12/31/2018. Inclusion criteria were suspected sepsis, SBP >90 mm Hg and MAP >65 mm Hg, and first lactate ≥4 mmol/L within 6 hours of ED arrival. Exclusion criteria included pressors within 6 hours and no outpatient bicarb in 12 months. The primary outcome was pressor requirement within 6-24 hours from arrival. Covariates, including anion gap, history of chronic kidney disease (CKD), 24-hour fluid administration, and lactate level were explored.


Results: A total of 16,713 patients were identified, 13,457 were excluded, and 3,256 were included. Median age was 72 years and 45% were female. Median initial lactate for no-drop vs drop was 4.5 mmol/L (IQR 4.2, 5.1) vs 5.1 mmol/L (IQR 4.4, 6.5), p< 0.001, and median total fluids in 24 hours for no-drop vs drop was 3.3 L (IQR 2.0, 4.5) vs 3.5L (IQR 2.0, 5.0), p< 0.001. In total, 335 (10.3%) needed pressors at 6-24 hours. Of patients with no-drop in bicarb, 7.2% required pressors vs 10.8% in those with a drop (RR 0.67, 95% CI 0.48,0.93). Drop in bicarb has a sensitivity of 89% (95% CI 85%, 92%) and an NPV of 93% (95% CI 90%, 95%) for pressors. In multivariate analysis, the RR for pressors for no-drop vs drop was 0.81 (95% CI 0.58, 1.13), and the RR for 24-hour fluids was 2.16 (95% CI 1.75-2.68).


Conclusions: Most normotensive, septic, ED patients with elevated lactate did not require pressors. After adjusting for 24-hour fluids, no-drop in bicarb did not significantly protect against need for pressors. However, it may still be useful in triaging at the time of admission.