Nine Years in Neurology
When I was in medical school studying anatomy, the professor spent two weeks on the cerebral cortex, but only gave the brainstem three class sessions. At the time, I thought this was reasonable. The cerebral cortex handles thinking, language, and memory—it's complex. The brainstem is just a relay station where signals pass through.
After nine years in neurology, I know: the brainstem is not a relay station.
In 2016, I admitted a patient: male, 47 years old. That morning, he woke up unable to open his left eye, and he kept veering to the right when walking. The CT showed nothing wrong. MRI revealed a brainstem infarction—the lesion was 8 millimeters in diameter. Eight millimeters. Smaller than a pinky fingernail. This man spent three weeks in the ICU.
An infarction of the same size in the frontal lobe might cause no symptoms at all. In the brainstem, it could cost you your life.
The brainstem volume is approximately 25-30 cubic centimeters. The entire brain is 1200-1400 cubic centimeters. The brainstem accounts for about 2%.
Packed into this 2% are: nuclei for 10 of the 12 cranial nerve pairs, the cardiovascular center, the respiratory center, the vomiting center, core nodes of the arousal system, and the mandatory pathways for all ascending and descending tracts.
If a certain area of the cerebral cortex is damaged, other areas might compensate. The brainstem has no compensation. Damage to the medullary respiratory center means a ventilator. Damage to the pontine micturition center means a catheter.
During ward rounds with my interns, I ask them how long the brainstem is. Most can't answer. The answer is 7-8 centimeters. From the midbrain to the junction of the medulla and spinal cord—that's how short it is.
Textbooks divide the brainstem into three segments: midbrain, pons, and medulla. Clear boundaries, beautiful diagrams.
In actual imaging interpretation: when a lesion straddles the junction of the pons and medulla, where do you localize it? Do you write pons or medulla? Some reports write "pontomedulary junction." On exams, writing this would cost you points. In clinical practice, it's very common.
Textbooks say the brainstem has a "reticular formation" that controls arousal and consciousness. It's drawn as a mesh of dotted lines. When I was in graduate school, I thought this was just a conceptual thing, a functional area without clear boundaries.
It's not. The reticular formation contains dozens of nuclei, each with its own name, each with specific coordinates. The locus coeruleus, the raphe nuclei, the pedunculopontine tegmental nucleus. Undergraduate textbooks just don't cover these. Even if they did, they wouldn't be on the exam.
In my second year of residency, I admitted a patient with hypersomnia—sleeping fourteen or fifteen hours and impossible to wake. After completing all the tests, we finally localized it to the reticular activating system in the brainstem tegmentum. At the time, I stared at those dotted lines in the textbook. Useless. I spent three days going through the literature before I figured it out.
Imaging diagnosis of brainstem lesions. CT is basically useless. The skull base artifact is too heavy, and the brainstem is in that location—on CT, it's just a blur.
Conventional 3.0T MRI, slice thickness 5 millimeters. The brainstem diameter is about 3 centimeters. Six slices and you're done scanning. A 2-millimeter lesion might fall right between two slices—volume averaging effect—invisible.
I've seen patients whose MRI came back normal, but clinically their symptoms were clearly a brainstem syndrome. Went to another hospital for thin-slice scanning, 1-millimeter slice thickness, and there was the lesion.
Tertiary hospitals generally do thin-slice scanning in radiology now. County-level hospitals not necessarily. The equipment has this capability, but scanning takes longer, and when there are many patients, they might not use it.
The doctors reading the scans also vary. Brainstem anatomy is complex—which nucleus damage corresponds to which symptoms requires memorizing too much. I've seen reports that say "abnormal signal in brainstem" with no further localization. Reports like that are of limited clinical help.
Time window for brainstem stroke. Textbooks say the thrombolysis time window for acute ischemic stroke is 4.5 hours. That's data from cerebral hemisphere studies. Data specifically for the brainstem is scarce.
The brainstem's blood supply mainly comes from the vertebrobasilar system. The hemodynamics of this system are different from the internal carotid system. Some brainstem infarctions are progressive—symptoms are mild in the first few hours, then worsen progressively.
A case that stayed with me:
This kind of progressive brainstem infarction—if you follow the standard time window, you might miss the intervention window. If you don't follow the standard, there's not enough evidence to support your decision. Many decisions are made in gray zones.
A graduate student I mentored asked me if the brainstem is good for publishing papers. I said it depends on what you want to publish. High-impact papers are hard—too few cases, can't do large-sample studies. Low-impact papers are easy—case reports, rare syndromes, one patient can make one paper.
He asked me if the brainstem is worth pursuing. I didn't answer directly. I asked him if he could read literature about a 7-centimeter structure without getting frustrated.