Volatile induction/maintenance of anaesthesia with sevoflurane increases jugular venous oxygen saturation and lumbar cerebrospinal fluid pressure in patients undergoing craniotomy

Abstract

Sevoflurane is a widely used volatile agent in neuroanaesthesia, and its nonirritant properties on the respiratory tract make sevoflurane suitable for volatile induction/maintenance of anaesthesia (VIMA). In the current trial, we investigate the effects of sevoflurane VIMA on lumbar cerebral fluid pressure (LCSFP) and jugular venous oxygen partial pressure in patients undergoing craniotomy for supratentorial intracranial surgery under bispectral index monitoring, determine jugular bulb venous oxygen saturation (SjvO2) and calculate cerebral perfusion pressure (CPP), arteriojugular venous oxygen content difference and cerebral oxygen extraction rate (COER) at different time points during the operation. Moreover, we compare these with total intravenous anaesthesia (TIVA) in patients receiving remifentanil and propofol continuous infusion. Twenty-six adult patients undergoing elective craniotomy were randomly allocated to either remifentanil-propofol (group TIVA) or sevoflurane (group VIMA). Mean arterial blood pressure, heart rate, LCSFP, CPP, SjvO2, arteriojugular venous oxygen content difference and COER were compared in the two groups at different time points under bispectral monitoring. Heart rate, mean arterial blood pressure, saturation of peripheral oxygen, BIS value and LCSFP were recorded at the time points including intubation (Tintub), 5 min (T5'int) and 15 min (T15'int) after intubation, the beginning of surgery (Tsurg), 5 min (T5m), 0.5 h (T0.5h), 1 h (T1h), 1.5 h (T1.5h), 2 h (T2h) and 2.5 h (T2.5h) after incision, the end of surgery (Tend), extubation (Textub) and 30 min (T30'end) after surgery. Arterial blood gas and jugular venous bulb blood gas analysis were performed at the time points of T0, T5'int, T5m, T1h, T2h, Tend, 15 min after operation (T15'end) and T30'end. LCSFP was increased above baseline levels in group VIMA and decreased in group TIVA from Tintub to T5m, and decreased to similar levels from T0.5h to T30'end in both groups. Mean arterial blood pressure was decreased to similar levels from T0.5h to Tend and returned to baseline levels at T30'end in both groups, and was significantly higher in group VIMA than in group TIVA at T5'int, T15'int, Tsurg and T5m. CPP was decreased to similar levels between the groups from T0.5h to Tend and returned to baseline levels at T30'end in both groups, and was significantly higher at Tintub, T5'int and T15'int in group VIMA than in group TIVA. Jugular venous oxygen partial pressure and SjvO2 were increased above baseline level from T5'int to T30'end in group VIMA, and decreased from T5'int to Tend, and returned to baseline level at T30'end in group TIVA. Arteriojugular venous oxygen content difference and COER were decreased from T5'int to T30'end in group VIMA, and increased above the baseline level from T5'int to Tend, and returned to the baseline level at T30'end in group TIVA. Sevoflurane VIMA increases SjvO2, lumbar CSF pressure and CPP, and decreases COER, which suggested that VIMA could be a better choice for patients with the risk of cerebral hypoperfusion or insufficient oxygen delivery. Propofol-based TIVA is associated with decreased SjvO2, LCSFP and CPP, and increased COER, and it might be suitable for patients with increased intracranial pressure.