Predictive performance of central venous oxygen saturation, lactate, base excess, and alactic base excess for intradialytic hypotension

Abstract

Background: Intradialytic hypotension (IDH) remains one of the most frequent and clinically significant complications of hemodialysis, associated with adverse cardiovascular outcomes. Conventional hemodynamic monitoring provides limited predictive accuracy, and the role of metabolic biomarkers as early indicators of tissue hypoperfusion in this setting is not well established. This study aimed to evaluate the predictive value of central venous oxygen saturation (ScvO₂), lactate, base excess (BE), and alactic base excess (ABE) in identifying patients at risk of IDH during maintenance hemodialysis. Methods: We conducted an exploratory observational study at the Hemodialysis Unit of the National Medical Center of Bajío (IMSS) between 2020 and 2021. Adult patients with central venous catheter access were eligible. Venous blood samples were collected at baseline, 90 min, and the end of dialysis to determine ScvO₂, BE, lactate, and ABE. IDH was defined per K/DOQI criteria. Diagnostic performance was assessed using paired comparisons, receiver operating characteristic (ROC) curve analysis, and logistic regression models. Results: Of 176 screened patients, 50 were included (mean age 49 ± 16.7 years; 54% male; mean APACHE II score 12.2 ± 4.6). IDH occurred most frequently within the first 90 min. A decline in mean arterial pressure > 10 mmHg was the most sensitive diagnostic criterion. Among biomarkers, BE and ABE showed significant dynamic changes, while lactate decreased modestly during dialysis (p < 0.05). ScvO₂ exhibited a non-significant mid-session decline. ROC analysis revealed that ScvO₂ best predicted early IDH (AUC 0.70, 95% CI 0.45–0.82), whereas baseline lactate ≥ 1.6 mmol/L demonstrated the highest accuracy for late-onset IDH (AUC 0.80, 95% CI 0.54–0.89). Logistic regression confirmed elevated baseline lactate as an independent predictor of late IDH (OR 3.46, 95% CI 1.21–9.84, p < 0.05). BE and ABE showed limited predictive value. Conclusions: Baseline lactate and dynamic changes in ScvO₂ provide complementary diagnostic information for predicting IDH in maintenance hemodialysis, while ABE may offer exploratory insights into metabolic compensation. Incorporating metabolic biomarkers into conventional monitoring could enhance early risk stratification and support individualized dialysis prescriptions. Larger multicenter studies are needed to validate these findings and establish diagnostic thresholds.