 Many older patients experience a kind of brain fog following surgery in which cognition is impaired from lingering effects of anesthesia. It's temporary but can still be disruptive. Now there is early evidence that one way to prevent such problems is to automate more aspects of anesthetic management during surgery. That's the preliminary finding from a new randomized controlled trial appearing in the journal Anesthesiology. Researchers at a hospital in Belgium tested whether automating three aspects of anesthetic management, anesthetic depth, cardiac blood flow, and protective lung ventilation improved performance on cognitive tests post-op compared to when an anesthesiologist is in manual control. Going in, the idea was that machines could do an even better job than humans at keeping parameters within the recommended ranges and this might lead patients to have less post-op cognitive impairment. The team recruited 90 patients 60 years or older with intermediate or high-risk elective non-cardiac surgeries. Half had anesthetic management controlled by a physician and half used closed-loop systems to automate management with an anesthesiologist on hand to override the machine if necessary. The day before surgery all patients took a variety of cognitive tests including the Montreal Cognitive Assessment, a 30-item global cognition test often used to detect mild cognitive impairment and dementia. In the week following the surgery patients were reassessed and once again at three months after surgery. The researchers found that patients with the automated group had on average no change in their cognition scores from before and one week after surgery. The manual group on the other hand had a one-point decrease which was statistically significant and this persisted at three months post-surgery. None of the other cognitive tests showed differences between groups. As expected the automated group had more anesthetic modifications per hour and ultimately received less anesthetic. Patients in the automated group spent more time within the target range for anesthetic depth than controls. Interestingly there was a correlation between the amount of time spent in very deep anesthesia and a decrease in the cognition score. The results suggested automated anesthetic management may help with delayed neurocognitive recovery possibly by avoiding very deep anesthesia. The study however cannot distinguish between the effects of the three controller systems. Future studies are needed with more patients and with those at higher risk of cognitive impairment.