Collaborators: Emery Brown (MIT Picower Institute, IMES / MGH Anesthesiology), Nancy Kopell (Boston University), Sydney Cash (MGH Neurology)
How do anesthetic drugs produce states of altered consciousness? How can we measure these brain states clinically to ensure that patients are receiving the right amount of sedative or anesthetic drugs? In this project we focus on the neural circuit mechanisms of anesthesia, using a multi-scale, multimodal approach spanning human studies in volunteers and neurosurgical patients, computational modeling, and animal models. Our goals are to uncover the secrets of the unconscious and anesthetized brain, and to enable precise, personalized anesthetic care for all patients.
The majority of patients who receive general anesthesia for surgery are older than 65 years of age. Older patients have a much higher risk of developing post-operative cognitive problems, including delirium and cognitive dysfunction. Why are older patients more susceptible to these cognitive problems? Are there things we could do, such as improved brain monitoring or alternative anesthetic regimens, to reduce this cognitive risk? In this project we use cutting edge neuroimaging, high-density electroencephalogram (EEG), and neurocognitive assessments to characterize brain function and cognitive outcomes in older adults receiving general anesthesia for surgery. We aim to 1) identify markers of brain health in older patients that might predict poor post-operative cognitive outcomes, 2) develop methods for EEG-guided anesthesia care tailored to older patients, and 3) develop alternative approaches for anesthesia care in older patients designed to preserve cognitive function.
Alzheimer’s disease AD affects tens of millions of patients worldwide. The disease processes underlying AD are thought to begin many years before the appearance of symptoms. Magnetic resonance imaging (MRI) and positron emission tomography (PET) are the “gold-standard” brain imaging tools currently used to characterize the pre-clinical stages of AD, but these methods are very expensive, making them impractical for large-scale testing. Could we develop alternatives that provide just as much or more information, but at a fraction of the cost? The electroencephalogram (EEG) provides rich information about brain dynamics. We know that these dynamics change during aging and dementia, but the relationship between these changes and underlying disease processes is poorly understood. With new EEG analysis methods, many of which are being developed in this laboratory, we can localize the sources of these brain dynamics at a resolution comparable to MRI and PET. In this project we aim to 1) characterize the relationship between the EEG and neuroimaging (PET and MRI) markers of brain health, 2) develop low cost, sensitive tools using the EEG to identify pre-clinical AD and track its progression.
The opioid epidemic is one of the largest public health crises in modern history. In this research, we are developing new tools to monitor and control nociception (pain) in the operating room (OR) and intensive care unit (ICU), as well as new tools to identify drug effects in substance overdose patients. By addressing nociception more precisely in the OR and ICU, we can reduce post-operative pain, which in turn reduces post-operative opioid requirements and the risk of opioid dependence. In addition, we are applying our knowledge of drug mechanisms and brain dynamics to help first-responders and emergency room staff identify drug effects in overdose patients, which can help healthcare providers take better care of substance overdose patients.
Collaborators: Takao Hensch (Harvard), Charles Nelson (Children’s Hospital Boston), Charles Berde (Children’s Hospital Boston), Laura Cornelissen (Children’s Hospital Boston), Erik Shank (MGH Anesthesia), Paul Firth (MGH Anesthesia), Rob Williams (UVM Anesthesiology), Emmett Whitaker (UVM Anesthesiology), Jerry Chao (Einstein Montefiore Anesthesiology), Choon Looi Bong (KK Women’s and Children’s Hospital, Singapore)
Each year in the United States, millions of children undergo general anesthesia or sedation for essential surgical and medical procedures. Parents of these children often worry about the effects of anesthetic exposure on their child’s development. Empirically, anesthesiologists know that brief (< 1 hour) anesthetic exposures have no effect on later neurocognitive development, but they know far less about the effects of longer exposures, repeated exposures, or lengthy sedation during intensive care. Few anesthesiologists appreciate the fact that anesthetic drugs act through neural mechanisms that can directly influence brain development. How could we be more certain at fundamental level that anesthetic exposure would not influence a child’s neurodevelopment, particularly if that child requires multiple surgeries or lengthy sedation? Could we provide children with a more precise and tailored anesthetic using the EEG to monitor their brain? Finally, how might we use our knowledge of anesthetic mechanisms to learn about brain development in children? We are applying our knowledge of neuroscience, engineering, and anesthetic mechanisms to address all of these questions.
Funding: NIH, NSF
Neuroscience generates some of the most interesting and complex data in science, yet prevailing methods for analyzing these data remain for the most part unsophisticated. This significantly limits the insights that might be derived from these data. Moreover, the absence of appropriate analysis tools limits the potential clinical benefits of neuroscience research, since clinical translation demands a higher standard of precision and reliability. To address this problem, if only for our own data, my lab has placed a heavy emphasis on developing novel statistical signal processing methods for neuroscience. We have focused on the EEG source localization problem, in which we use high-density scalp measurements to estimate underlying cerebral currents. Our approach is to use biophysical and functional neuroanatomic principles to guide the solution to this problem. We have discovered that neurophysiologically-principled models of sparsity, connectivity, and most importantly dynamics can significantly expand the realm of what is possible using source localization, improving spatial resolution significantly, and even enabling localization of subcortical activity. We are currently developing novel signal processing tools for EEG and local field potentials that more efficiently represent brain oscillations, significantly improving statistical efficiency. Although this work requires substantial time and effort, it is worth it, because in the end we gain so much more insight from our hard-earned data, and improve our ability to apply those insights clinically.
Our education modules provide free CME credit (https://eegforanesthesia.iars.org)
Collaborators: Emery Brown (MIT Picower Institute, IMES / MGH Anesthesiology)
Clinical education is one of the most important ways that we can improve patient care. Our fundamental insights about brain dynamics during anesthesia make it possible for anesthesiologists and nurse anesthetists to provide more precise, personalized anesthesia care for their patients. Our lab engages in clinical education both at the local level via in-person and online education at MGH, and at the international level through online initiatives with the IARS and lecture presentations at major international meetings.