One of neuroscience’s foundational experiments wasn’t performed in a Nobel laureate’s lab, but occurred in a railyard in 1848 when an accidental explosion sent a tamping iron through 25 year-old Phineas Gage’s forehead. Gage survived, but those studying his history detailed distinct personality changes resulting from the accident. He went from even-tempered to impulsive and profane. The case is likely the earliest—and most famous—of using a “lesion” to link a damaged brain region to its function. In the ensuing decades, to study the brain was to study lesions. Lesion cases fed most of the era’s knowledge of the brain.
One might think that modern neuroscience, with its immense toolkit of experimental techniques, no longer needs lesions like Gage’s to parse the brain’s inner workings. Lesion studies, though, seem to be having a revival. A new method called lesion network mapping is clearing the cobwebs off the lesion study and uniting it with modern brain connectivity data. The results are revealing surprising associations between brain regions and disorders.
Thankfully, most lesions aren’t a tamping iron through the forehead. Strokes, hemorrhages, or tumors make up most lesion cases. 19th century neurologists like Paul Broca made foundational discoveries by studying patients with peculiar symptoms resulting from these common neurological insults. Broca and his contemporaries synthesized a theory of the brain from lesions: that the brain is segmented. Different regions control different functions. Lesion studies lend a lawyerly logic to the brain: if region X is destroyed and function Y no longer occurs, then region X must control function Y.
This logic, though, is a bit misleading. No single brain region can really control any function. The modern view of the brain is that individual functions rely on a network of interconnected brain regions working in concert. Thus, modern neuroscience views individual lesion cases as imperfect, uncontrolled experiments of nature that don’t necessarily speak to how a network controls a brain function. This point becomes obvious when researchers pool, or meta-analyze, lesion data. When looking at all of the published lesion cases for a given condition—say parkinsonism—researchers see that the lesions that cause this condition don’t occur in just one region. Lesions seemingly all over the brain cause parkinsonism and other conditions. This fact, along with the emergence of elegant experimental tools have pushed lesion studies to the sidelines of neuroscience. Some researchers, though, are attempting to revive the relevance of lesion studies—both for neurology and psychiatry. The authors of a new study published in Brain use the disparate locations of lesions to their advantage—in this case to better understand the neuroanatomy of parkinsonism.
Parkinsonism is a grouping of symptoms affecting movement. It consists of slowed movement, rigid musculature, and tremor. The most common cause of these symptoms is Parkinson’s disease, where dopamine-producing cells in the substantia nigra are progressively lost. Nigral cell loss in Parkinson’s disease occurs through a slow degenerative process that is still poorly understood. But parkinsonism is possible without nigral degeneration, and notably can occur following a sudden lesion like a stroke or hemorrhage. Patients with lesion-induced parkinsonism aren’t diagnosed with Parkinson’s disease, exactly, but their slowed movement, rigid musculature and tremor are nearly identical to those with “classical” Parkinson’s disease. The study compiled 29 published cases of lesion-induced parkinsonism. The lesions did not all occur in the same region, and surprisingly most were not in the substantia nigra.
The authors hypothesized that the parkinsonism-causing lesions, despite occurring in disparate brain structures, disrupt common connectivity networks in the brain. To test this, the authors overlaid these lesion locations on a map of the brain known as the connectome—a structural map of region to region connectivity derived from functional MRI data. With the lesions applied to the connectome, the authors were able to identify networks—or tracks of connectivity—that the lesions disrupted.
Each of the 29 lesions sat within several different networks, which is to be expected as the brain is a rich tangle of connectivity. But the authors saw that 28 out of 29 cases affected networks that connected through a small, sheet-like structure called the claustrum. The claustrum is rarely discussed in the field of movement control or Parkinson’s disease, and is generally understudied.
An important aspect of the study is that none of the 29 lesions were to the claustrum, itself. It took the combination of the lesions and the connectome to identify the claustrum as a structure of importance for parkinsonism.
The claustrum coincidentally appears as a rest stop, so to speak, on the network maps of almost all the lesion cases. But is this just a coincidence or is it important? To address the claustrum’s importance to parkinsonism, the authors turned to patients with the more common, degenerative form of Parkinson’s disease that had deep brain stimulators implanted in their brain. Deep brain stimulation is a treatment of last resort for Parkinson’s disease and doesn’t yield universal improvement. In most cases, stimulating electrodes are implanted into a region called the subthalamic nucleus. The precise location within the subthalamic nucleus varies patient to patient. The authors examined the precise location of deep brain stimulators within Parkinson’s disease brains and overlaid those locations onto the connectome. They saw that when electrode locations were within networks that flowed through the claustrum—presumably altering claustrum activity—patients saw better results from deep brain stimulation. This result argues that claustrum activity plays a critical role in generating parkinsonian movement. Also, altering that activity provides relief from parkinsonism.
19th century lesion studies were framed by the question, “which region controls which function”. Decades of neuroscience have reframed a more nuanced question, “which regions are important to which functions?” Lesion network mapping empowers lesion studies to rigorously answer this newer question. To patient communities, however, the question has always been “can this finding help us?” In the case of the claustrum and Parkinson’s disease, only time will tell. Targeting treatments—like deep brain stimulation—to the claustrum, though, may be a helpful advance for those with Parkinson’s disease.