Anesthesia during asleep-awake craniotomy should provide adequate analgesia and sedation whereas permitting language testing. In this work, we used the analysis of heart rate variability (HRV) to quantify the sympatho-vagal balance and better evaluate patient's stress response during asleep-awake craniotomy. Patients admitted to our hospital for tumor resection with language testing were studied (n=21, age range: 22 to 53 y ). Heart rate and systolic arterial blood pressure were collected at 5 time points: T1: preanesthesia; T2: dura mater opening; T3: cortical mapping; T4: subcortical mapping; T5: dura mater suturing. Patients were anesthetized with propofol/remifentanil infusion and ventilated via laryngeal mask during T2, but were awakened for language testing at T3 and T4, and resedated with remifentanil during T5. At each time point, HRV was analyzed by power spectrum analysis: overall variance, very low frequency (VLF), low frequency (LF) and high frequency (HF) powers, and LF/HF ratio (an index of prevalence of sympathetic over parasympathetic tone) were derived. A significant increase in both heart rate and systolic arterial blood pressure was observed from time point T3 through T5 (P<0.05, compared with T1). HRV analysis revealed that the LF/HF ratio progressively increased to reach values during T4 that were significantly higher than preanesthesia values (P<0.05). During T5, LF/HF ratio returned to preanesthesia level. HRV analysis confirmed the presence of moderate intraoperative stress response, indicating a significant increase in the LF/HF ratio during the awake phases. This information might help in tailoring the protocol and the duration of awake phase according to the individual autonomic response.