Background: The prevalence of postoperative cognitive disturbance, coupled with growing in vitro, cell, and animal evidence suggesting anesthetic effects on neurodegeneration, calls for additional study of the interaction between surgical care and Alzheimer neuropathology. The authors studied human cerebrospinal fluid (CSF) biomarkers during surgery.
Methods: Eleven patients undergoing idiopathic nasal CSF leak correction were admitted to this Institutional Review Board-approved study. Lumbar subarachnoid catheters were placed before the procedure. Anesthesia was total intravenous propofol or remifentanil or inhalational sevoflurane, depending on provider choice. CSF samples were taken after catheter placement (base), at procedure end (0 h), and then at 6, 24, and 48 h. CSF was analyzed using xMAP Luminex immunoassay (Luminex, Austin, TX).
Results: Of the 11 patients (age range, 53 ± 6 yr), 8 were women; 4 received intravenous anesthesia, 6 sevoflurane, and 1 mixed. Procedures lasted 6.4 ± 2 h. Mean CSF amyloid-β(1-42) remained unchanged, but total-tau and phosphorylated-tau181P increased progressively until at least 48 h. Total-tau, phosphorylated-tau, or amyloid-β(1-42) concentrations were not different between anesthetic groups. CSF interleukin-10, S100Beta, and tumor necrosis factor α were increased similarly in both anesthetic groups at 24 h, but interleukin-6 was increased more in the inhalational group.
Conclusion: These data indicate a robust neuroinflammatory response, including not only the usual markers (interleukin-6, tumor necrosis factor α, interleukin-10), but also S100Beta and tau, markers of injury. The total-tau/amyloid-β(1-42) ratio increased in a pattern consistent with Alzheimer disease, largely because of an increase in total-tau rather than a decline in amyloid-β(1-42). The differences in CSF interleukin-6 concentrations suggest that anesthetic management may make a difference in neuroinflammatory response.