Association of jugular venous oxygen saturation and mean arterial pressure variability with dynamic cerebral autoregulation and outcomes in sepsis-associated encephalopathy
摘要
Sepsis-associated encephalopathy (SAE) is a frequent and clinically important manifestation of sepsis-related acute brain dysfunction. Impaired cerebral autoregulation and altered cerebral oxygen supply-demand balance are considered relevant pathophysiological features, but the temporally aligned and statistically adjusted relationships among jugular venous oxygen saturation (SjvO2), mean arterial pressure (MAP) variability, and dynamic cerebral autoregulation (dCA) remain insufficiently defined.
ObjectiveThis retrospective study examined the associations between SjvO2, MAP variability, and dCA indices in neurocritical care patients with SAE and evaluated whether these monitoring variables were associated with impaired dCA and short-term outcomes.
MethodsWe retrospectively analyzed 200 adult neurocritical care and intensive care patients with SAE selected from an indication-based advanced neuromonitoring registry between January 2022 and December 2025. Jugular bulb catheterization was performed for clinical indications rather than as universal screening. SjvO2 monitoring, invasive arterial pressure monitoring, and transcranial Doppler-based dCA assessment were matched within the same early neuromonitoring window after SAE recognition. MAP variability was quantified using the coefficient of variation (MAP-CV) over the 24-hour period containing the first valid dCA session, and SjvO2 was derived from the matched 60-minute monitoring epoch rather than pooled ICU-stay data. Dynamic cerebral autoregulation was assessed by transfer function analysis and the mean flow index (Mx). Multivariable logistic and Cox regression models were adjusted for prespecified illness-severity and physiological covariates.
ResultsThe mean SjvO2 was 62.4 ± 11.8%, with 34.5% of patients demonstrating abnormal values (< 55% or > 75%). MAP-CV was 14.2 ± 5.6%, and impaired dCA (Mx > 0.3) was observed in 58.0% of patients. SjvO2 was negatively correlated with Mx (r=-0.487, p < 0.001) and positively correlated with phase difference (r = 0.412, p < 0.001). MAP-CV was positively correlated with Mx (r = 0.523, p < 0.001) and negatively correlated with phase difference (r=-0.394, p < 0.001). Multivariable analysis identified SjvO2 < 55% (OR = 3.24, 95%CI: 1.78–5.89, p < 0.001), MAP-CV > 15% (OR = 2.87, 95%CI: 1.56–5.28, p = 0.001), and SOFA score (OR = 1.18, 95%CI: 1.06–1.32, p = 0.003) as independent predictors of impaired dCA. Patients with both abnormal SjvO2 and high MAP-CV had higher 28-day mortality than those with normal SjvO2 and low/moderate MAP-CV (51.0% vs. 22.1%, p < 0.001). In a simultaneous Cox model including impaired dCA, the combined abnormal SjvO2/high MAP-CV phenotype remained associated with 28-day mortality (adjusted HR = 2.54, 95%CI: 1.42–4.55, p = 0.002).
ConclusionsLower SjvO2 and greater MAP variability were associated with impaired dCA in SAE patients. The combined assessment of cerebral oxygenation and MAP variability may help identify patients with higher neurological and mortality risk within an observational risk-stratification framework. These findings are hypothesis-generating and require prospective validation before being used to define individualized hemodynamic targets or treatment effects.