copyright 1999

INTRODUCTION

Since there are already a plethora of resources on neuroanatomy, many of which you will have ready access to, why burden you with another? The attempt to put together a brief treatment of human neuroanatomy which tries to put forth in simple terms the major structural and functional features of the system, with emphasis on clinical relevance and stressing methods of organizing material to aid recall and prevent common pitfalls of misunderstanding, is probably similar to the aim behind, for example, Goldberg's Clinical Neuroanatomy Made Ridiculously Simple or Barrow's Guide to Neurological Assessment. (The latter, however, also includes and focuses on the procedures used to look for findings.) Since these texts (and others) are done well, a felt need to produce another is probably to some extent, parochial--looking for a resource that seems to "just fit" our curricular needs--written at just the right level of detail, stressing what we consider important, reflecting the particular "mix" of neuroanatomy, neurophysiology, neuroradiology, and clinical relevance being aimed for in our curriculum. Beyond that, the major differences from other resources I am familiar with will be to have you participate actively throughout--much as when using programmed self-learning texts, such as Sidman and Sidman's Neuroanatomy: A Programmed Text and to have you review what you've already learned over and over--each time a new functional system is added--by working relevant problems.

In a sense, I am at cross-purposes in writing this workbook. On the one hand, it is meant to be an overview of neuroanatomy only, and so should have a strong structural focus. Yet functional neuroanatomy is probably the main component of the Neuro block, which is constructed on the premise that neuroanatomy in isolation, like neuro-physiology, neuropharmacology, neurochemistry, etc., in isolation, is not very worthwhile clinically. Since it is the integration that is not only worthwhile but also most difficult, I am continually tempted to weave in material that is not strictly neuroanatomy and to leave out complicating structural details that do not seem useful. For now, I will simply suggest that as soon as possible, you should be tying your anatomy into the other disciplines necessary to handle clinical problems. This will allow you to begin judging for yourself what level of neuroanatomical detail you will need to know. Indeed, assuming that you will use neuroanatomy as a physician, the natural starting point will be a finding, not a structure. Thus, a natural progression in your studying might be to begin with findings, immediately ascertain and learn how to do the procedures necessary to demonstrate the findings, then study the mechanisms underlying the findings--including anatomy, physiology, biochemistry, etc. If working in reverse, trying to "make sense" of neuroanatomy, for example, then similarly become involved in the relevant physiology, testing procedures, and findings (normal and abnormal) as soon as possible. In this regard, you will probably want to keep your set of Neuro Clinical Skills modules (and perhaps a copy of Barrows, Guide to Neurological Assessment) handy, along with your black bag and a willing (?) subject (girl and boy friends, spouses, and children are historical favorites).

The faculty here has been impressed with how much better students seem to learn when they assume an active role--continuously asking questions about material they encounter in lectures, books, or patients and forcing themselves to make drawings, examine patients, etc.--rather than simply listening or reading, hoping that when they find the "magic resource" or have gone over the material enough times or in enough different ways, the material will suddenly fall together, all make sense, and be at their command. How often one hears a student say--if only someone would put all the brainstem structures in one clear diagram or 3-D model (despite availability of such resources!) or if only someone would tell us what's important to know (despite many people trying to do so!), etc. The bottom line is--the material is vast and complexly interwoven enough that you must use it to learn it. And use it means working with it until you feel comfortable that you know what's important and why, until you could make your own clear diagrams or 3-D models of the brainstem and defend why you chose the structures you did and put them where you did! At that point, one might say that you've "digested" the material, and it's the digestion that's critical. A student sometimes seems to think he can digest an integrative lecture by simply reading the lecture notes over till memorized, or that she can digest a diagram or model by passively memorizing what it looks like. No, memorization is not digestion. Memorization is duplication. One might just as well take a photograph of the notes or diagram and put the photo in his pocket. No matter how big a bite is taken, as long as the student relates it to nothing else, it is isolated. To assimilate material into your way of thinking about problems that require the information, you must reorganize it-- personalize it by forcing it to encounter and to make sense with the rest of your relevant knowledge. Only then will the material have been digested and become something that is uniquely yours. Good luck!

NEUROANATOMICAL TERMINOLOGY: SOME COMMENTS

Even if functional neuroanatomy had been "discovered" by a close-knit team of educationalists, very concerned about making what they found out easy to learn, the number of structures and functions alone would make for a hefty list of new terms to be learned. When you add to the picture the changes in structures that occur during development, embryologic and later, the procedures needed to examine the structures and functions for normality, and the various abnormalities of structure and function that are found, you are well on your way to learning a whole new language. Now, top this all off with the fact that many structures, functions, procedures, findings, etc., have names that are not descriptive, and may, in fact, have more than one name, sometimes not quite synonymous: and there you have it! The wonderful world of neurologic jargon!

Don't underestimate this hurdle. The sooner you become comfortable using these terms--speaking them and writing them in meaningful context, as well as understanding them when you hear them or see them--the sooner you can start meaningfully thinking about observations and problems that involve them. How can we make this necessary process easier for you? Hard to say, but here are some suggestions!

1. Immerse yourself in the material (we should be able to help you with this), but don't let yourself drown! When you don't understand a term, ask or look it up. If the new terms initially are so plentiful that you can't look them all up, concentrate on learning first the ones that seem necessary to understand the gist of what's going on. If you don't understand the words, it's hard to understand the events they are describing. And if you don't understand the events, it's hard to enjoy working with them to solve problems and learn more.

2. Don't be afraid to spend a good bit of time simply learning new words, especially ones that seem obviously important to you or that pop up repeatedly. I would estimate that all of you will encounter at least 500 new words a week, probably closer to 1000 for the first few weeks. Certainly 1-2 hours a day on vocabulary alone would not be unreasonable during the first few weeks. Obviously, how much time you spend can't be dictated too exactly, as it will depend on your background, your capabilities, and your interest.

3. If and when you do work on vocabulary alone, try to find ways that make your study more fun and help to tie the words together. For example, try not to look up long lists of unrelated terms one at a time. In Objective 1 of NEURO ANAT 1, we've tried to help by listing some pairs and triplets of related terms; we've also tried to list closely related pairs and triplets together. In Figure 1, I've put together some associational "families" of terms to extend the pair-triplet notion. Obviously, I had some particular relationships in mind as I generated these pairs, triplets, and families, but remember that my relationships should not be the focus here--yours should. In other words, don't worry if you don't see some of the relationships right away, or if you aren't sure that a relationship you see is the one I had in mind, or if you think other pairs, triplets or familial linkings make better sense. It might be useful for you to make your own list of pairs of terms which you find yourself often confusing, so you can review them periodically. Your goal should be to learn the terms in a way that makes sense to you and that you think will help you to remember them. By the way, if you come up with some new way of "playing with terms" that you find useful, please share it with me. Perhaps others would also find it useful.

STRUCTURE OF THE NERVOUS SYSTEM: AN OVERVIEW

On to the structures themselves! This workbook will make very heavy use of diagrams and make no use of actual photographs. Consequently, make sure that you use your cadavers and fixed brains, as well as various models of the nervous system and anatomy texts, often enough that you are in fact "picturing" structures accurately as you think about them, their locations, and their functions.

Figure 2 is a diagrammatic frontal section of the nervous system, showing many of the major divisions and structures. Remember it is only accurate enough to allow meaningful labelling of these major divisions and structures. There is no attempt to draw it to scale. Most of the terms used to label Figure 2 (and Figure 3and Figure 4 as well) are "biggies" and therefore should be learned soon. You'll be meeting them over and over again.

Note that in terms of gross structures the nervous system is bilaterally symmetrical around the midsagittal plane. This means that there are "duplicates" of many structures--one on each side of the midline. This can be confusing for the beginning student, for we often talk of the paired structures as if there is only one--e.g., the optic nerve (cranial nerve II), the frontal-lobe (of the cerebral cortex), the caudate nucleus (one of the basal ganglia), the corticospinal tract (carrying information from cerebral cortex to spinal cord). This is just a convention, where it is assumed that the reader or listener knows that there are really two such structures. Some names, however, refer to only one structure on both sides of the body. Cerebrum is an example; other examples are brain stem, cerebellum, and spinal cord. As you learn the name of each structure, be sure to learn whether there are two of them or only one. To make it more confusing, there are a few cases where clinicians and neuroanatomists use a name differently. For example, clinicians tend to use "thalamus" to refer to thalamic structures in both cerebral hemispheres--i.e., there is only one thalamus. Neuroanatomists, however, often refer to a thalamus in each cerebral hemisphere- -i.e., there are two thalami (plural of thalamus). We will speak of only one thalamus, since you are clinicians-in-training. But now, if you run into the term "thalami" in a neuroanatomy text, it shouldn't surprise or bother you.

The "cerebrum" is what in layman's terms might be called the "brain." It is divided into right and left cerebral hemispheres. Its outer surface (think of it as a peel or rind) is called the cerebral cortex. Since it is loaded with cell bodies and dendrites, which are unmyelinated, the cerebral cortex looks gray--it is an example of gray matter. Other major areas of gray matter in the cerebral hemispheres are the basal ganglia and thalamus. Since the rest of the cerebral hemispheres (excepting the ventricles--cavities filled with cerebrospinal fluid--to be mentioned later) consists of axons, most of them myelinated, these areas look white--they are examples of white matter. The axons in these white areas act as wires, carrying information (neural impulses) from cell bodies in some portion of gray matter to cell bodies (and/or their dendrites) in some other portion of gray matter. Students frequently confuse the terms "cerebral" and "cortical." "Cerebral" refers to the entire cerebrum, so "decerebrate" refers to a condition in which the entire cerebrum is thought to be functionally disconnected from the rest of the nervous system. "Cortical," on the other hand, refers to only the cerebral cortex, so "decorticate" refers to a condition in which the cerebral cortex (but not the entire cerebrum) is thought to be functionally disconnected from the rest of the nervous system. (An illustration of the postures typical of decerebrate and decorticate rigidity are given on p. 674 of Curtis, Jacobson, and Marcus, An Introduction to the Neurosciences.) As a side note, "cortical" could theoretically refer to cerebellar cortex as well (see below); but as an isolated term, it always refers to cerebral cortex only.

Caudal to the cerebrum is the brain stem. The brain stem connects the cerebrum to the spinal cord, which begins by definition where the brain stem ends--at the foremen magnum. The brain stem is usually broken down into three major subdivisions. The one adjacent to the cerebrum (thalamus) is the midbrain, also known as the mesencephalon. Caudal to the midbrain is the pons. And caudal to the pons, joining it with the spinal cord, is the medulla--also called medulla oblongata. Although clinicians tend to consider only the midbrain, pons, and medulla as divisions of the brainstem, neuroanatomists often include the thalamus (thalami) as part of the brainstem. Keep this in mind to prevent confusion as you read neuroanatomy texts. The brain stem, like the cerebrum, is made up of areas of white matter and areas of gray matter. However, unlike the cerebrum, there is no outer covering of gray matter. The gray matter, instead, is found internally, grouped into a lot of distinct areas called nuclei (plural of nucleus).

Lateral and caudal to the brain stem is the last major portion of the central nervous system within the skull--the cerebellum. Cerebellum means "little cerebrum," and like the cerebrum, the cerebellum is mostly white matter inside (except for some deep nuclei--to be discussed later), surrounded by gray-looking cortex at the surface. Notice the similarity between the terms cerebrum and cerebellum, also between the terms cerebral cortex and cerebellar cortex. As you might imagine, it's easy to use the wrong term or hear the wrong term without even realizing it. If you're sensitive to this potential problem, you will probably have no difficulty with it.

The only portion of the central nervous system below the head is the spinal cord. Rostrally the spinal cord is defined to end at the foremen magnum, where the brain stem begins. Caudally, it ends in the adult at about the level of the twelfth thoracic or first lumbar vertebra. (See NEURO EMBR 1 for the embryologic development leading to this placement and NEURO ANAT 3 for more information on the relevance of this placement for doing spinal taps, also known as lumbar punctures.)

As a quick review, then, we've now covered the major divisions of the central nervous system (CNS). They are: the cerebral hemispheres (including the basal ganglia and thalamus), the brain stem (including midbrain, pons, and medulla), the cerebellum, and spinal cord. There now, that wasn't so bad, was it?

The peripheral nervous system (PNS) is much simpler to describe than the CNS, for it consists of two sets of structures only: 12 pairs of cranial nerves (and their roots, branches, and ganglia) and 31 pairs of spinal nerves (and their roots, branches, and ganglia). By definition, spinal nerves attach to the spinal cord, and cranial nerves attach to a part of the CNS inside the cranium.

Having gone over the CNS and PNS, we have covered the entire nervous system. There are no other neural structures left. But wait a minute, you may say. What about the autonomic nervous system, the limbic system, the sensory system, the extrapyramidal system, etc.? Good point. The answer is: any additional subdivisions of the nervous system that you hear about are simply different sets of structures within the CNS and/or PNS. Usually, a particular set of structures is singled out and given a particular name because someone has thought that they form a functional unit. For example, the autonomic nervous system (also called vegetative nervous system) was thought to regulate vital internal functions, such as pulse and respiratory rate, blood pressure, and digestion. The limbic system was thought to regulate emotional and complex innate behaviors, such as those associated with aggression, eating and drinking, and mating. And so on. As we progress further, we will describe some of these functional sets of structures. But for now, the message should be--if you've named the major components of the CNS and PNS (as we already have), you have named the major components of the entire nervous system.

So much for Figure 2. Now let's look at Figure 3, which is a schematic diagram of a cross-section (transverse section) through the spinal cord. The spinal cord, like the brain stem, has its white matter (myelinated neural processes) on the outside and its gray matter (cell bodies and unmyelinated neural processes) inside. Unlike the brain stem, however, the internal gray matter is continuous (not broken up into a lot of discrete nuclei) and has a characteristic shape, somewhat like an "H". The upper (dorsal) limbs of the H are called the dorsal or posterior horns and the lower (ventral) limbs of the H are called the ventral or anterior horns. In the thoracic, lumbar, and sacral cord (not the cervical), there is a lump of gray matter protruding laterally off the middle of the H; it's called the intermediolateral cell column and will be mentioned later, when we talk about the autonomic nervous system. Just as the gray matter is divided mainly into the dorsal and ventral horns, the white matter is divided basically into three sets of funiculi (plural of funiculus = a small cord or bundle). On the two sides of the midline between the dorsal horns are the two (right and left) dorsal funiculi. On each side, between the ipsilateral dorsal and ventral horns is one of the two (right or left) lateral funiculi. And on the two sides of the midline between the ventral horns are the two (right and left) ventral funiculi.

Figure 4 is a cross-section through one lateral half of a spinal cord, with a spinal nerve attached. Each spinal nerve is formed from the union of a dorsal nerve root and a ventral nerve root. The dorsal nerve root itself comes off the dorsal spinal cord as a series of small rootless. Similarly, the ventral nerve root comes off the ventral spinal cord as a series of small rootless. The nerve fibers running in the dorsal root are all sensory; the nerve fibers running in the ventral root are all motor. Thus, if a patient had a lesion near the dorsal spinal cord that destroyed several adjacent dorsal roots on one side, we would expect him to have sensory loss on that side of the body, but no motor problems. Similarly, if a lesion wiped out only several adjacent ventral roots on one side, we would expect to find motor loss on that side of the body, but no sensory deficits. Once the dorsal and ventral roots of a spinal nerve join, however, motor and sensory fibers tend to run together, so that typically a patient with functional impairment of one or more spinal nerves (or both dorsal and ventral nerve roots) has both motor and sensory problems, with similar distributions in the body.

To avoid confusion, we should probably discuss right away various classifications of sensory systems that you will be encountering. One of the simplest and most clear-cut classifications distinguishes among exteroception, proprioception, and interoception. Exteroceptors convey information about the external environment; examples are receptors in the eyes, ears, nose, and skin. Proprioceptors convey information about location and movement of different parts of the body; examples are receptors in muscles, tendons, and joints. Interoceptors convey information about internal organs, such as the heart, lungs, blood vessels, stomach and intestines, etc.; often, these sensations are poorly differentiated as to quality and location, but deep pressure and deep pain are common types of interoceptive experience. A second classification distinguishes between general sensory modalities and the special senses. This classification completely cries-crosses that of extero-, proprio-, and interoception. Within exteroception,, touch, superficial pain and pressure, and temperature--all conveyed from receptors in the skin--are general senses; whereas vision, audition, and olfaction are special senses. Within proprioception, information conveyed from receptors in joints, ligaments, muscles and fascia is general; whereas information on orientation and movement of the head conveyed from the vestibular system is special. Within interoception, deep pain and deep pressure are general senses; whereas taste is a special sense. As a rule, if a type of sensation is conveyed from various points of the body, it's called general; whereas sensations which arise only from a specific organ or tissue, which is usually specialized in some way, are called special. Clinically, another common category of sensation is somatosensation (also called the somesthetic system, some = body, esthesia = capacity for sensation). Somatosensation is a subclass of the general sensations, which includes only the general exteroceptive senses and general conscious proprioception (i.e., sense of position and sense of movement conveyed from receptors in joints, ligaments, and fascia).

The primary (first-order) general sensory neurons begin at their distal end in the form of sensory endings or adjacent to specialized receptor cells. Once an action potential is begun in this region, it must travel into the CNS before the information it represents is processed in any way. Two specializations help to hasten the speed with which sensory impulses reach the CNS. First, the neural process from the sensory endings (or receptor cell) to the cell body, which is embryologically a dendrite, has been structurally made into an axon: it can carry action potentials and many of the processes are even myelinated to allow much faster velocities (see NEURO PHYS 2 to learn how myelination speeds up velocities of action potentials). Indeed, these dendrites are not even called dendrites. The fibers running in nerves are all called axons. Second, the sensory first-order neurons have become unipolar: that is, the cell body has migrated to one side, allowing the embryological dendrite to join directly to the embryological axon (see Figure 5). This means that the action potentials do not have to be transformed into local potentials on the cell body, and then summate adequately before a new action potential is generated in the embryological axon (see NEURO PHAR 2 to learn about local potentials and summation). Instead, each single action potential can speed on its way, having bypassed the cell body. The cell bodies of all first-order sensory neurons traveling in a given spinal nerve are found in a special bulge on the dorsal root of that nerve: the bulge is called the dorsal root ganglion (ganglion = swelling). Note that in these sensory ganglia, there are no synapses. [NOTE: Unlike spinal nerves, cranial nerves do not usually have clearly separate sensory and motor roots--the trigeminal (cranial nerve V) is an exception. However, except for the olfactory and optic nerves (don't worry about them now), cranial nerves that have general sensory fibers do have sensory ganglia. In most cases, these ganglia are also outside (though close to) the CNS and contain cell bodies of the unipolar first-order neurons--i.e., no synapses. The two exceptions are as follows. (1) The first-order sensory neurons of the stato-acoustic nerve (cranial nerve VIII) are bipolar, though the cell bodies are still segregated into sensory (spiral and vestibular) ganglia. For these neurons, another specialization has substituted for the lack of unipolarity: the cell bodies have become myelinated, so that action potentials in the embryologic dendrites can simply "jump" across the cell bodies to the embryologic axons (see NEURO PHYS 2 to learn about saltatory conduction). (2) The cell bodies of the unipolar first-order sensory neurons associated with proprioception in muscles innervated by the trigeminal nerve (cranial nerve V) make up a small nucleus in the midbrain--called the mesencephalic nucleus of V. This is the only true ganglion in the CNS; all the rest are in the PNS. The "basal ganglia" don't count, for they are really nuclei, despite their name.

Many motor neurons that send their axons out of the CNS through the ventral root of a spinal nerve innervate striated muscle fibers. They have their cell bodies in the ipsilateral ventral (anterior) horn at the same level of the spinal cord where the ventral root containing their axons comes off. Because of the location of their cell bodies, these neurons are sometimes called anterior horn cells. They are the only motor neurons that innervate striated muscle fibers. A more common name for these neurons is lower motor neurons (or lower motoneurons--LMNs). Since they are called lower, you might expect that there are upper motor neurons (UMNs) as well. There are. We won't discuss them till a little later, but for now, just realize that a lot of CNS neurons with their cell bodies above the spinal cord send axons down into the spinal cord to influence the firing of the lower motor neurons. [Don't get confused into thinking that all motor neurons above the spinal cord are upper motor neurons. Not true! Motor neurons whose axons travel in cranial nerves to innervate striated muscles are also lower motor neurons, even though their cell bodies reside in the brain stem. Remember, then, that you can have LMN lesions in the brain stem as well as in the spinal cord and peripheral nerves; likewise, you can have upper motor neuron lesions in the spinal cord (the axons of UMNs must travel in the spinal cord to reach the appropriate LMNs) as well as in the brain stem and cerebrum. These points will become important a bit later, when we describe the differential findings in LMN and UMN lesions. Although there are a variety of neurons from various parts of the nervous system that do influence the activity of striated muscles, the only way they can do so is to act upon the lower motor neurons. For this reason, the lower motor neurons have been referred to as the "final common pathway." In other words, an LMN may be receiving impulses from some neurons "telling it" to fire faster and impulses from other neurons "telling it" to fire more slowly, but it is the net effect of all these impulses on the LMN that constitutes its "decision" as to how rapidly it will fire. Another way to assess the significance of LMNs would be to sever them. If all LMNs innervating a muscle were severed or destroyed, the CNS would have no way to make the muscle contract: the muscle would be denervated. [As a side-note, sensory fibers from the muscle might still be intact (see NEURO PHYS 3 and 4 to learn about sensory innervation of striated muscles), but the muscle would still undergo denervation atrophy and display denervation hypersensitivity (see NEURO PHYS 2 and NEURO PHAR 9 to learn about these).

There are other motor neurons that send their axons out of the CNS through the ventral root of spinal nerves, to innervate smooth muscles and glands. Although their cell bodies are also in the ipsilateral spinal cord at the same level where the ventral root containing their axons comes off, the cell bodies are in the intermediolateral cell column (see Figure 3), not the anterior horn. These neurons are part of a very important functional neural system, called the autonomic (vegetative) nervous system. Though autonomic fibers (axons) are found all over the body, the autonomic system is particularly important in regulating a variety of vital functions--including those served by the heart (e.g., pulse rate), lungs (e.g., respiratory rate), stomach (e.g., rate of acid production), intestines (e.g., force and rate of peristaltic movements), blood vessels (e.g., blood pressure), urinary bladder (e.g., ability to urinate), and reproductive organs (e.g., erection and ejaculation in the male), as well as many others.

In the broadest sense, the autonomic nervous system can be (and is) considered to have both afferent (sensory) and efferent (motor) components, the afferents being the sensory neurons innervating the variety of smooth muscle-containing and gland-containing structures (mostly, internal organs) served by autonomic efferents. These autonomic (vegetative) afferents, in fact, make up the interoceptors, carrying sensory information about internal organs, as opposed to the proprioceptors and exteroceptors mentioned earlier. In the narrower sense, however, the autonomic nervous system is often spoken of as strictly an efferent system. For example, when you read or hear that the autonomic system is basically a two-neuron system, consisting of preganglionic neurons and postganglionic neurons (as in NEURO ANAT 4), be aware that both the preganglionics and postganglionics are efferent, neurons. Similarly, when you hear or read about autonomic neurotransmitters (as in NEURO PHAR 3), be aware that it's the neurotransmitters of the preganglionics and postganglionics only, i.e., efferents, that are meant. In the brief overview to be given here, also, it is only the efferents that will be described.

The autonomic nervous system is divided into two major components: the sympathetics and the parasympathetics. These two components, though having some characteristics in common, show major differences in both anatomy and function. Both the sympathetic and parasympathetic systems consist of both preganglionic and postganglionic neurons. The cell bodies of all preganglionic neurons are in the CNS. At the spinal cord level, the preganglionic cell bodies are in the intermediolateral cell columns; in the brain stem, they are in distinct nuclei. Preganglionic cell bodies of sympathetic and parasympathetic systems are not found together, however. Preganglionic cell bodies of the sympathetic system are found only in the thoracic and lumbar segments of the spinal cord; while those of the parasympathetic system are found only in the brain stem and in the sacral spinal cord. Note that there are no preganglionic cell bodies in the cervical spinal cord; thus there are no intermediolateral cell columns in the cervical cord. Although some organs of the body may be innervated by sympathetics only or by parasympathetics only, most autonomically innervated structures receive both sympathetic and parasympathetic fibers. This is an important point, as in any given structure, the effects of sympathetic firing tend to be opposite to the effects of parasympathetic firing. As just one example of this, increased sympathetic firing speeds the heart rate, while increased parasympathetic firing slows it down. The sympathetic system has been called the flight or fight system, which prepares an animal under stress to take appropriate action. The parasympathetic system, on the other hand, seems to promote functions associated with non- stressful daily existence, such as digestion. (See NEURO PHAR 3 to learn more about autonomic functions.)

But if the parasympathetics originate only in the brain stem and sacral cord, how do they come to innervate all the structures in between? And if the sympathetics originate only in the thoracolumbar cord, how do they reach the head (and neck) and lower body? Let's take the parasympathetics first. The tenth cranial nerve (the vagus nerve) carries a large number of autonomic fibers (all are parasympathetic, of course). Although it is a cranial nerve, it travels down through the neck into the thorax and abdomen, innervating a variety of structures down to (but not including) the sacral level. The sympathetics, for their part, move toward the head and sacral regions in two bundles of sympathetic fibers, one on each side of the vertebral column, called the paravertebral sympathetic chains. At various points along these chains, not only at the ends, bunches of sympathetic fibers leave to accompany blood vessels, which carry them to their organs of destination. Since it is the sympathetics that tend to be distributed along blood vessels, it may not be surprising that the main autonomic innervation of the limbs, which terminates primarily in the blood vessels themselves (especially the arterioles) and in the skin (especially sweat glands--sudomotor fibers--and smooth muscle that erects the body hair-- pilomotor fibers) is sympathetic.

Next we will consider the autonomic ganglia and postganglionic neurons. All autonomic ganglia are found in the peripheral nervous system. In them, the axonal endings of preganglionic neurons synapse with the dendrites/cell bodies of postganglionic neurons. Note, therefore, that autonomic ganglia, unlike sensory ganglia, contain synapses. The parasympathetic ganglia tend to be found near or in the organ to be innervated by the postganglionic neurons. Thus, the parasympathetic preganglionics (as in the vagus nerves) tend to be long; whereas the parasympathetic postganglionics tend to be very short. Since the parasympathetic ganglia tend to occur toward the peripheral end of the system, after the preganglionics have already broken up into rather small bundles going to the various organs, the ganglia are relatively small, often microscopic. Discussion of the sympathetic ganglia and sympathetic postganglionic neurons takes us back to Figure 4.

Although Figure 4 was meant to represent any level of the spinal cord, what you've already read about the autonomic system lets you know that it is really accurate only for the thoracolumbar spinal cord. Since the cervical cord has no cell bodies of preganglionic autonomic neurons, it would not have the intermediolateral cell column shown in Figure 4. And although the sacral cord does have cell bodies of preganglionic autonomic neurons, they are parasympathetics, so would not have axons going into the chain of sympathetic ganglia lying along the vertebral column (paravertebral ganglia), as does the axon of the representative preganglionic neuron in Figure 4. Notice in Figure 4 that each thoracic and lumbar spinal nerve is attached to the paravertebral ganglia by two small bundles of fibers, called rami (plural of ramus, meaning "branch"). One of the rami is called a white ramus, the other is called a gray ramus. Why? Well, the preganglionic sympathetic fibers course from the spinal cord through the white ramus to the paravertebral chain. Since all preganglionic fibers are myelinated (this is true of parasympathetics, too!), the ramus in which they run appears whitish. The postganglionic sympathetic fibers, however, whose cell bodies are found in the paravertebral ganglia, leave the paravertebral chain through the gray ramus, to enter the spinal nerve and go peripherally. Since all postganglionic fibers are nonmyelinated (this is true of parasympathetics, also), the ramus in which they run appears grayish. Let me hasten to point out, however, that not all sympathetic fibers synapse in the paravertebral ganglion nearest the spinal nerve in which they exit. Some go up or down the paravertebral chain to other paravertebral ganglia before synapsing; others go to one of several large sympathetic ganglia anterior to the vertebral column (prevertebral ganglia), and a few are even found in small terminal ganglia near the organ to be innervated (e.g., in the urinary bladder). This concludes our brief overview of the autonomic nervous system (see NEURO ANAT 4 and NEURO PHAR 3 to learn more about it).

SOME LONG-TRACT SYSTEMS: THE FOUNDATION

Now that we've finished our first quick journey through the nervous system, let's go through it again. But this time, we'll concentrate on function of structures, particularly functions that are often tested clinically to find out what, if anything, is wrong with the nervous system, and where the problem is located. The process of finding out where problems exist in the body is called anatomical diagnosis. As you will soon discover, the process of anatomical diagnosis is central to the clinical neurosciences. Although there are some more or less "silent" areas in the nervous system, where damage may not give rise to abnormal findings during the taking of the history (symptoms) or during the physical examination (signs), damage in most areas does produce some abnormal symptoms and/or signs. Because of the way structures subserving various functions are organized in the nervous system, the findings associated with damage to a particular area of the nervous system are often unique to that area. Often, by specifying abnormal findings carefully and knowing the functional anatomy of the nervous system, a physician can locate neural damage to within a centimeter or less. Although the degree of sophistication attained by many clinical neuroscientists is awe- inspiring, one can achieve considerable skill at neuroanatomical diagnosis rather quickly by focusing on major organizing facts and principles. We will try to maintain that focus on this, our second, journey through the nervous system.

We will start our journey by consideration of three sets of structures, each of which is centrally important clinically, as damage to it will produce a specific constellation of abnormal findings. As a short-hand convention, we will refer to each of these three sets of structures as a "long-tract" system, to signify that each set stretches through a considerable portion of the nervous system. What these three sets have in common, then, is that damage at a variety of locations in the nervous system may disrupt their functions! Two of the long-tract systems are general afferent (sensory)--the "lateral spinothalamic" system and the "dorsal columns" system. The third system is efferent (motor)--the "corticospinal (corticobulbar)" system.

Let's consider the lateral spinothalamic system first (Figure 6 and Figure 7). This system carries information on pain and temperature. Clinically, these modalities are tested on the skin, light pin-pricks being used to measure ability to sense superficial pain and warm/cool objects being used to measure ability to sense temperature (see NEURO HxPx 5 to learn more about the clinical testing). In all examinations of general sensory systems, the patient's eyes should be closed.

Just from information you've learned so far, the initial course of the lateral spinothalamic system should not surprise you. As all general sensory systems do, the lateral spinothalamic begins with first-order neurons, whose unipolar cell bodies reside in a sensory ganglion. The impulses conveyed by these neurons begin at the distal ends, which may be specialized themselves or be adjacent to specialized receptor cells, and pass proximally right through the sensory ganglion and into the CNS.

Having entered the ipsilateral dorsal horn of the spinal cord, the axons of the first-order neurons end (without ascending or descending) by synapsing on dendrites and/or cell bodies of second-order neurons in the dorsal horn. The axons of these second-order neurons begin to ascend and at the same time to move towards the anterior white commissure (Figure 6), where they cross to attain a position in the ventral half of the contralateral lateral funiculus. It takes about 2-3 spinal segments of ascension before all the axons of second-order neurons associated with first-order lateral spinothalamic neurons in a given spinal nerve have crossed. For example, for first-order neurons in the left tenth thoracic nerve, some axons of associated second-order neurons will be crossing the midline at the T10 level, but others will be crossing at T9 or T8. This is clinically important to remember. For example, if you tested a patient and found her unable to sense superficial pain and temperature on the left side from the foot on up to the level innervated by the tenth thoracic nerve, would damage to the right half of the cord at the T10 spinal level be consistent with the findings? No! To wipe out pain and temperature on the left up to the level innervated by the tenth thoracic nerve would require that the right cord be damaged at about T7-T8 (or above, if it didn't wipe out the entire pathway). If this doesn't make sense to you, go over the above material on the lateral spinothalamic system again.

Since you will meet neural fibers crossing the midsagittal plane (midline) again and again in various functional systems, let's lay out two general definitions now. Several anatomically distinct regions in the CNS which join together structures on the two sides (left and right) are called commissures (commissura = a joint or a joining together). Often, a commissure is the site of many neural fibers crossing the midline, as in the case of the anterior white commissure of the spinal cord. If a bundle of fibers (actually one bundle on either side) crosses the midline while simultaneously moving anteriorly or posteriorly, the two bundles together tend to form an "X" shape as viewed in cross-section. Since "decussate" means to cut or cross in the form of an X, the crossing fibers are said to "decussate" and the place of crossing is called a "decussation." The decussation of the second-order axons in the lateral spinothalamic system, which takes place in the anterior white commissure, is an example (Figure 6). You will soon meet several other important decussations. Although you may at times hear the word decussate used loosely to mean simply "cross the midline," purists will try to maintain the original distinction.

The actual "lateral spinothalamic tract" itself is situated in the ventral half of the lateral funiculus of the spinal cord. It originates in the lower sacral cord and runs all the way up the cord to the brain stem. Since it is gaining fibers from more and more of the body as it ascends, the lateral spinothalamic tract becomes increasingly larger as it approaches the medulla.

Since this is the first tract we have encountered, let's digress briefly to note some general features of tracts. We'll start with the names. Fortunately, most names of tracts have some descriptive aspects, making it easier to recall them correctly (assuming you know something about the tracts, that is). Many names of tracts tell you both their origins and destinations--in that order. In that regard, some common combining forms are cortico- (= cerebral cortex), spino- (= spinal cord), bulbo- (= brain stem), thalamo- (= thalamus), and cerebello- (= cerebellum). The name "spinothalamic," for example, means that the tract originates in the spinal cord and travels to the thalamus. The "lateral" in front of "spino-thalamic" indicates that this tract runs in the lateral funiculus of the spinal cord. Another spinothalamic tract, called the "anterior" spinothalamic, runs in the anterior funiculus of the spinal cord. Besides giving information about origin, location, and destination, which is clinically quite useful, some names give information about shape of the tract as it appears on cross-section--not particularly helpful. Examples are the fasciculus cuneatus (cuneatus = wedge-shaped) and internal arcuate fibers (arcuate = shaped like a bow). Least helpful are terms that are basically synonyms for "tract," such as bundle, fasciculus, and lemniscus.

Since most tracts were identified as anatomical entities long before much was known of their function, the correlation between tracts and functional pathways is often far from 1:1. For example, you can see from Figure 6 and Figure 7 that the lateral spinothalamic tract makes up only a portion of the lateral spinothalamic system. Although a given tract usually contains only uninterrupted axons (i.e., no synapses), it often does not even include the entire axons. In the case of the lateral spinothalamic tract, for example, the axons of the second- order neurons travel from one dorsal horn, through the anterior white commissure to the contralateral lateral funiculus, before they officially enter the tract. Also, in some cases, there are nuclei essentially within a tract, so that the tract doesn't end at one location, but rather ends partially in stages as some of its fibers end in synapses at each nucleus. (The descending (spinal) tract of the trigeminal is an example, to be discussed later.)

While fibers within a given tract (whether called a tract, bundle, fasciculus, lemniscus or whatever!) tend to run together and are often visible on cross- sections as separate structures, the fibers may at times be separated from each other by intervening structures, sometimes widely so. As you might imagine, this can have important clinical consequences. Where fibers of a tract are running close together, a relatively small lesion may completely destroy the tract and its functions; whereas in areas where the fibers are diffusely spread out, a lesion of similar size may result in almost no apparent functional deficit. The corticospinal system, an effector system to be described soon, provides an example here, for a cut through the dorsal part of one lateral funiculus in the cervical cord (1-2 millimeters) can completely destroy ability to move the ipsilateral arm, leg, or body below the neck, whereas a lesion near the cerebral cortex would have to extend over much of the precentral gyrus and paracentral lobule (8-10 centimeters!) to cause the same deficit.

A last point to be mentioned is that fibers in several important tracts are organized somatotopically--i.e., fibers within the tract are grouped physically according to the part of the body they influence (some = body, topos = a place). For example, a common organization for tracts while in the cerebrum and rostral brain stem is to have fibers associated with the feet and legs at one end, fibers associated with the arms and trunk in the middle, and fibers associated with the head and neck at the opposite end. Often, where the fibers associated with a particular part of the body will be located within a tract can be reasoned out, so that you don't have to memorize the information.

For example, think about how the lateral spino-thalamic tract is formed as it ascends (Figure 8). Since at any level of the cord, the axons just joining the tract come to its medial side first, having just decussated, one could say that it "makes sense" to have the fibers already in the tract "pushed" further laterally and dorsally. Thus, fibers associated with the perineum, which had already joined the tract in the sacral cord, would be found most laterally, while fibers associated with the legs, trunk, arms, and neck would, respectively, be located more and more medially. Fortunately, this organization not only makes sense, but is true. The lateral spinothalamic tract will also be used to illustrate the possible clinical significance of somatotopic organization. If you remember, we determined a few pages back where a lesion might produce loss of ability to sense superficial pain and temperature on the left side from the toes on up to the level of the trunk innervated by the tenth thoracic nerve. At that point, we decided on a lesion in the ventral half of the right lateral funiculus at T7-T8, wiping out the entire right lateral spinothalamic tract at that level. Now that you know about the somatotopic organization of the tract, however, you might see another possibility. Right! A lesion higher up in the spinal cord (e.g., at the level in Figure 8) may have mimicked T-7 damage by destroying only the dorso-lateral part of the right lateral spinothalamic tract.

When the lateral spinothalamic tract goes from the upper cervical spinal cord into the medulla, the only major change that occurs is in its name (Figure 7). When entering the medulla from below, fibers of the anterior and lateral spinothalamic tracts merge, so they are now jointly called by the same name--the spinothalamic tract. The spinothalamic tract courses on up through the medulla, pons, and midbrain to the thalamus, where its axons end by synapsing in the ipsilateral ventral posterolateral nucleus (the VPL).

As you learn more about the nervous system, you will discover that several different sensory systems that send information to the cerebral cortex do so via the thalamus, where the cell bodies of the neurons whose axons will actually reach the cortex are segregated into distinct nuclei. These nuclei are collectively known as thalamic relay nuclei. Each sensory system has its own, and the VPL is the relay nucleus for the somatosensory system, which includes the lateral spinothalamic. Since the axons in the spinothalamic tract belong to second-order neurons, the cell bodies in the VPL belong to third-order neurons. Because axons from cells in thalamic relay nuclei tend to fan out as they approach the cortex, they are called thalamic radiations. Because the axons coming from the VPL are associated with somatosensation, they are called the somatosensory radiations. During the first part of their trip, as they pass around the basal ganglia, they are within a structure called the internal capsule. The internal capsule contains essentially all neural fibers connecting the cerebral cortex with other parts of the nervous system. It is very important clinically and you'll be learning more about it soon. After passing the basal ganglia, the somatosensory radiations become much more diffuse as they fan out and intermingle with a variety of other fiber systems. In this location, they form a small part of the myriad fibers radiating out in all directions to the cortex. These myriad fibers are jointly called the corona radiate (corona = crown, radiate = radiating). The part of the corona radiate formed by the somatosensory radiations ends up synapsing largely in the ipsilateral post-central gyrus (and posterior paracentral lobule) of the parietal lobe of the cerebral cortex.

Don't worry about all the special names now. The most important facts to remember about the anatomy of the lateral spinothalamic system are that fibers in the system cross the midline within 1-3 spinal segments above their entry point into the cord, are somatotopically organized, and remain crossed (i.e., contralateral to their receptors) all the way up through the spinal cord, brain stem, and thalamus, to the cerebral cortex. As for function of the lateral spinothalamic system, remember two things. First, the system conveys information on pain and temperature, which are tested clinically with a pin and warm/cool objects, respectively. Second, lesions in the system up to and including the VPL in the thalamus can destroy all ability to sense temperature and pain in the relevant parts of the body, but lesions above the thalamus (somatosensory radiations or cortex destroy primarily the ability to tell where stimuli presented to the relevant parts of the body are located, even though the stimuli themselves may be correctly perceived as painful, hot, or cold. In other words, if you examined a patient and found that when you touched him on the right side of the body with a pin, he could feel it was sharp but couldn't locate very well where he'd been touched, you'd think of a lesion ... Where? Exactly! In the left somatosensory radiations or left somatosensory cortex.

Now that you've completed going over your first functional system, try posing some relevant clinical problems to yourself and working out the answers. To get you started, I'll give you a couple; but you should be able to come up with others yourself.

• PROBLEM 1. When doing a comprehensive neurologic examination on a 23-year-old black female, you find that she experiences pricks of a pin on her left hand and arm (C5-T1) as being dull, not sharp. You also find that she cannot distinguish warm and cool objects placed on that extremity. Everything else is normal. Where might her problem be located? Why?

• PROBLEM 2. A 72-year-old white male cannot sense superficial pain or temperature on his right side from his foot all the way up to the level of his umbilicus (T10). Everything else is normal. Where might his problem be located? Why?

• PROBLEM 3. A 32-year-old white female cannot sense superficial pain or temperature on either side from the nipple line (T4) up to the axilla (T2). Everything else is normal. Where might her problem be located? Why?

• PROBLEM 4. A 45-year-old white male cannot sense superficial pain or temperature on his left side from his foot all the way up to the lower border of his rib cage (T8) or on his right side from his umbilicus up to the lower border of his rib cage. Everything else is normal. Where might his problem be located? Why?

A quick comparison of Figure 9 with Figure 6 will show you that up to the point of entering the spinal cord through the dorsal roots, the dorsal columns system is organized just like the lateral spino-thalamic system. To repeat: As in all general sensory systems, the dorsal columns system begins with first-order neurons, whose unipolar cell bodies reside in a sensory ganglion. The impulses conveyed by these neurons begin at the distal ends, which may be specialized themselves or be adjacent to specialized receptor cells, and pass proximally right through the sensory ganglion and into the CNS. Unlike the lateral spinothalamic system, however, first-order neurons do not synapse soon after entering the cord. Instead, their axons go into the ipsilateral dorsal funiculus and ascend in it all the way up to the junction of the spinal cord with the medulla. Unfortunately, anatomists sometimes call the dorsal (posterior) horns, which consist of gray matter, the dorsal (posterior) columns. Be aware that whenever you encounter the term dorsal (posterior) columns clinically, it is the ascending long tracts making up most of each dorsal funiculus, i.e., white matter, that is meant.

The dorsal columns themselves are somatotopically organized. Given that as the dorsal columns ascend, axons newly joining them approach from the lateral side, where do you think that fibers from various parts of the body are found within the dorsal columns? If you guessed that the lower sacral fibers from the perineum, which enter first, become most medial, while fibers from the leg, trunk, arm, and neck lie, in that order, more and more lateral, then you guessed right! Although the dorsal columns on either side form a functional-whole, fibers that come from the perineum and leg (L1 on down) on one side are separated from the fibers originating in the ipsilateral trunk, arm, and neck by a deep narrow crevice. In neuroanatomy, a minor crevice, such as this one, is called a sulcus (= furrow, plural = sulci), a major crevice is a fissure. The name of the sulcus we've just described is not important. What is important is the unfortunate consequence of the sulcus' presence: the fibers on one side have a different name than the fibers on the other side. The fibers medial to the sulcus, emanating from the perineum and leg, are jointly called the fasciculus gracilis. Those fibers lateral to the sulcus, emanating from the trunk, arm, and neck, are jointly called the fasciculus cuneatus. When pairs of structures are arbitrarily assigned different names, it is usually very difficult to remember which name goes with which structure. In such situations, it's nice to have 2 mnemonic (= a device to aid your memory). Though I'm not sure it's worthwhile expending much effort to keep the fasciculus gracilis and fasciculus cuneatus straight, my own mnemonic is as follows. Gracilis and ground both start with "gr," cuneatus and ceiling with "c." The legs touch the ground, while the ceiling is higher, so the fasciculus gracilis is the medial column with fibers from leg and perineum, the fasciculus cuneatus is the lateral column with fibers from trunk, arm, and neck. A student suggested another mnemonic--namely, the gracilis muscle is in the leg. If you come up with a mnemonic you like better, let me know.

At the junction of the spinal cord with the medulla, the first-order axons in the dorsal columns end by synapsing on cell bodies/dendrites of ipsilateral second-order neurons. These synapses occur in a pair of nuclei: fibers in the fasciculus gracilis synapse in the nucleus gracilis, fibers in the fasciculus cuneatus synapse in the nucleus cuneatus. That seems almost too easy, doesn't it? Well, now the fun begins!

As you can see from Figure 10, the axons of the second-order neurons in the nucleus gracilis and nucleus cuneatus immediately cross the midline (decussate, remember?) before starting to ascend up through the medulla, pons, and midbrain. Unfortunately, name changes accompany these changes in position. As the axons are leaving the nuclei and crossing the midline, they are called internal arcuate fibers (because they are located far internal to the surface of the CNS and are bow-shaped). Once they begin to ascend, and until they reach their thalamic relay nucleus, the fibers are jointly called the medial lemniscus (because it is medially located during much of its ascent). Notice that axons leaving the nucleus gracilis do not have different names than axons leaving the nucleus cuneatus. From here on up, fibers associated with perineum, leg, trunk, arm, and neck all travel in structures with the same names. By comparing Figure 7 and Figure 10, you will see that once axons in the medial lemniscus end by synapsing on cell bodies of third- order neurons in the ipsilateral ventral posterolateral (VPL) nucleus of the thalamus, the rest of the structures are the same as in the lateral spinothalamic system. The axons of the third-order neurons form part of the somatosensory radiations, traveling first through the internal capsule and then through the corona radiate, finally synapsing largely in the ipsilateral post- central gyrus (and posterior paracentral lobule) of the parietal lobe of the cerebral cortex.

As in the case of the lateral spinothalamic system, don't worry about all of the special names of dorsal columns structures now. The most important facts to remember about the anatomy of the dorsal columns system are that fibers in the system are somato-topically organized, all of them decussate at the spino-medullary junction, and they remain crossed (i.e., contralateral to their receptors) all the way up through the brain stem and thalamus, to the cerebral cortex. As for function of the dorsal columns system, remember two things. First, the system conveys information on conscious proprioception, vibration, and discriminative touch, which are tested clinically by moving fingers and toes, using a tuning fork, and testing ability to localize tactile stimuli precisely. Second, lesions in the system up to and including the VPL in the thalamus can destroy all of the above--conscious proprioception, vibratory sense, and discriminative touch--in the relevant parts of the body (though crude touch will be spared in CNS lesions below the VPL); but lesions above the thalamus (somatosensory radiations or cortex will leave ability to perceive vibrations intact, though the patient may not be able to tell where the vibrating stimulus is located on the body.

This is a good time to pose some clinical problems involving only the dorsal columns system. Again, be sure to work through each problem, defending your answers. I'll only give you one this time.

• PROBLEM 5. A 43-year-old black male cannot feel vibrations in a tuning fork placed on bony prominences of his left foot and leg (e.g., lateral malleolus, shin, patella). He also cannot tell the position of his left toes, foot, or lower leg, nor the direction that you move them. Lastly, he cannot tell what numbers you "write" on his left foot or leg. (Other tests of discriminative touch also indicate deficits.) Ability to feel pain and temperature are normal. Where might his lesion be located? Why?

Let's start the process now. By comparing the anatomy and the function of the lateral spinothalamic and dorsal columns systems, you will already be able to refine your anatomical diagnoses considerably. Remember that until now, loss of sensations referable to just the lateral spinothalamic system or just the dorsal columns system, in part of the right leg, for example, could be caused by problems in the nerves to the leg themselves, in the spinal cord from the sacral level all the way up to the cervical, in the brain stem, or in the thalamus. (You already know how to tell if the problem is in the somatosensory radiations or cerebral cortex, right? If not, review briefly to refresh your memory.) Comparing Figure 7 and Figure 10 should help make the following functional-anatomical pattern clear. When you test sensations on the body (exclude the head for now), if the areas of loss for the two systems are about the same, then without consideration of other findings, you might expect that the problem is either peripheral (since the fibers for the two systems run together in the nerves) or in the cerebrum (since the fibers for the two systems are somatotopically integrated from the VPL nucleus on up to the cortex). Conversely, if you find an area of loss for one system only, it reflects damage in the spinal cord (where the lateral spinothalamic system is mostly contralateral, the dorsal columns system ipsilateral) or brain stem (where the two systems, though both now contralateral, don't run together). (I might point out, by the way, that even though the two systems cross at very different levels of the CNS, there is one striking similarity. Each of them crosses just beyond the cell bodies of the second-order neurons.)

Time to pose some more problems for yourself. Be sure to make some clinical (i.e., a patient presents with ... , so where's the problem?) and some anatomical (i.e., if there is damage to ..., what findings would the patient have?). I'll give you a couple.

• PROBLEM 6. A 65-year-old black male complains of dropping things with his left hand and stumbling when he walks. On examination, you find inability to locate stimuli accurately on the left side of his body, from his foot up to (and including) his head. He can, however, feel simple sensations-- touch, pain, temperature, vibration--on his left side. Where is his lesion located? Why?

• PROBLEM 7. A 16-year-old white male is carried into the emergency room, less than 1 hour after being in a motorcycle accident. He is alert, but in considerable pain. Neurologic findings include loss of position sense in both lower extremities, and loss of vibratory sense and two-point discrimination from the feet up to the umbilicus bilaterally. He also has loss of pain and temperature sensation in a 2-cm band around his trunk from the umbilicus down. No other sensory abnormalities. Where is his lesion located? Why?

Although the corticospinal-corticobulbar neurons and lower motor neurons all subserve function in striated muscles, damage to UMNs produces a different set of findings than damage to LMNs. (A few important exceptions will be given shortly.) To aid comparison of the two sets, some of the most commonly noted similarities and differences between them are listed in Table 1.

TABLE 1
Characteristic LMN Findings (flaccid paralysis)			Characteristic UMN Findings (spastic paralysis)
1.	paresis (i.e., partial paralysis) or paralysis (also called palsy)		paresis or paralysis (esp. of distal musculature--e.g., of fingers or toes)
2.	decreased superficial reflexes		decreased superficial reflexes
3.	decreased deep tendon reflexes (DTRs)		increased DTRs (also called hyperreflexia), possible clonus
4.	no pathologic reflexes		pathologic reflexes (e.g., Babinski and Hoffman)
5.	hypotonia (decreased tone)		hypertonia (increased tone)
6.	marked muscular atrophy (denervation atrophy)		mild muscular atrophy (disuse atrophy)
7.	fasciculations (and fibrillations)		no fasciculations (or fibrillations)

Your first reaction to this table might be--Oh, my God! Not only is it long, but it's full of unfamiliar terms, whose dictionary meanings are not very useful until you know how the relevant observations and maneuvers are actually done. It's not really as bad as it looks. Let's start by considering an LMN lesion, involving the muscles in, say, the right foot and leg. If you look down the list of characteristic LMN findings in Table 1, you'll notice that all of them (with the exception of fasciculations, but see below) are "absence" or "less than normal" findings. (Hardly surprising, since they reflect partial to complete absence of innervation to the muscles!)

1.	Paralysis is inability to contract muscles voluntarily. (If a patient can contract the affected muscles, but it's unclear whether the contractions are normal or weakened (paresis), you can do standardized tests of muscle strength--see NEURO HxPx 6 to learn more about them and the other clinical observations and maneuvers mentioned below.)
2-4.	Superficial reflexes are ones in which you stimulate skin receptors (e.g., with a light scratch) to make the muscles contract; deep tendon reflexes (also called myotasis or muscle stretch reflexes) are ones in which you stimulate muscle receptors by stretching the muscle slightly (e.g., by tapping the tendon with a rubber hammer) to make the muscles contract. Obviously, if direct innervation to the muscles is damaged, both kinds of reflexes will be decreased. Similarly, pathologic reflexes (i.e., presence of reflexes normally absent) cannot occur.
5.	Muscle tone is defined as the resistance to passive movement (e.g., to the examiner moving the foot, lower leg, etc., back and forth with the patient relaxed). Since a denervated muscle cannot resist movements that stretch it, the muscle is hypotonic. Active movement, by the way, is movement for which the patient provides the force. In tests of muscle strength, as in 1. above, it is strength of active movements (or of active isometric contractions) that is being measured.
6.	Atrophy (a = lack of, trophic = nutrition) means wasting, and muscle wasting can be observed grossly by comparing to the normal side of the body (if there is one). More precise measurement can be done with a tape measure. When a muscle is denervated, it diminishes greatly in size, presumably because the nerves to it somehow provide needed nutrients or necessary minimal activity.
7.	A fasciculation (from fasciculus = small bundle) is the simultaneous contraction (usually a very brief jerk) of a bundle of muscle fibers, all innervated by the same LMN. As you know, one LMN with all the muscle fibers it innervates is called a motor unit; and the muscle fibers in one motor unit are usually grouped together. When an LMN is damaged, changes in the membrane permeability to ions may cause action potentials, which occur independent of firing in neighboring LMNs, thus producing an isolated simultaneous twitch of the motor unit. Such fasciculations are often grossly visible to the eye. Once an LMN is dead, however, no further fasciculations occur; so fasciculations are a sign of damaged, yet viable LMNs. Fibrillations are similar autonomous firings, but in single denervated muscle fibers. Fibrillations are not grossly visible (except perhaps in the tongue), so demonstration is usually accomplished with electromyography--recording electrical activity in the muscle. Once a denervated fiber completely degenerates, it no longer fibrillates. Fibrillations, then, are a sign of denervated, yet viable muscle cells.

Now let's look at the characteristic UMN findings in Table 1. As you run down the list, notice that they can be divided into three groups--"less than normal" findings (paresis, decreased superficial reflexes, atrophy), a "normal" finding (no fasciculations or fibrillations), and "more than normal" or "release" findings (hypertonia, hyperreflexia and clonus, pathologic reflexes). As you can imagine, the "less than normal" findings, being similar to their LMN counterparts, aren't particularly helpful differentiators. Lack of fasciculations may also not be very helpful, as we noted above their absence in LMN lesions after the neurons are dead. Thus, it is the increased resistance to passive movement, the increased deep tendon reflexes, and the presence of usually absent reflexes that are the most useful indicators of UMN lesions.

You may be puzzled as to why a destructive lesion leads to an increase in activity, especially when it also produces paresis. This brings up two very important concepts, which you'll use repeatedly. First, if a set of neurons are inhibitory for a function, then their firing decreases the function and their electrical silence or destruction increases the function. If the function is normally inhibited by the neurons, then destruction of the neurons is said to "disinhibit" or "release" the function. In UMN lesions, for example, deep tendon reflexes (responses to stretch) are disinhibited, leading to hyperreflexia and hypertonia; and some pathologic reflexes (which are components of "normal" reflexes present in early infancy), such as the Babinski response, are released (see NEURO PHYS 4 and NEURO ANAT 17 to learn more about these reflexes).

But, you might say, what about the paresis and decreased superficial reflexes? If the UMNs are an inhibitory system, why don't we see more movement than normal and increased superficial reflexes? This brings up the second important concept. A set of neurons may influence more than one function; and if it does, some functions may be facilitated at the same time that others are inhibited. UMNs, for example, apparently facilitate voluntary movements and superficial reflexes, while inhibiting deep tendon and normal infantile reflexes. If our patient with weakness in the right foot and leg had an upper MN lesion, instead of a lower MN lesion, what would we find in addition to the paresis, the absent cremasteric reflex, some possible atrophy ("disuse" atrophy, which is less marked than "denervation" atrophy), and no fasciculations? There would be hypertonia on passively flexing and extending the foot and lower leg, hyperreflexia (e.g., of the patellar reflex and ankle jerk), and an extensor response of the large toe to planter stimulation (a Babinski response).

As in the case of the LMN lesion, however, the exact findings will depend on the time-course of the injury. Most importantly, you must realize that when a lot of damage to UMNs occurs suddenly, as, e.g., in a cerebrovascular accident (CVA = stroke) or trauma to the head or spine, the CNS goes into "shock," which masks the "release" phenomena. For example, if our patient with weakness in the right foot and leg had been fine until he suffered a stroke the previous day, we would not find hypertonia, hyperreflexia, and a Babinski response in the right lower extremity. Instead, we'd find hypotonia, hyporeflexia, and a normal planter response. Only over the course of days to weeks would his flaccid paralysis gradually be replaced by the signs of spastic paralysis.

After that long clinical-functional preamble, let's get back to the anatomy of the corticospinal-corticobulbar neurons. Just from the names, you can tell that they all have their cell bodies in the cerebral cortex, that the corticospinals end in the spinal cord, and that the corticobulbars end in the brain stem. Many of the cell bodies are found in the pre-central gyrus (and anterior paracentral lobule) of the frontal lobe (Figure 11). Let's deal with the corticospinal fibers first--those associated with function in muscles below the head. The axons coming from cell bodies in one cerebral hemisphere remain ipsilateral all the way to the medullospinal junction--traveling through the corona radiate somewhat diffusely, then becoming compact as they course through the internal capsule, and next passing through the midbrain pons, and medulla. Within the medulla, this corticospinal tract is somewhat pyramidal-shaped in cross-section, so has the alternate name of pyramidal tract. At the medullospinal junction, the vast majority of the corticospinal fibers decussate, in the so-called "pyramidal decussation," and form the lateral corticospinal tract, from which the various fibers are given off gradually and continuously down the length of the spinal cord (Figure 12). As you might imagine, if there is a lateral corticospinal tract, it suggests there is another corticospinal tract--and there is. It's called the anterior corticospinal tract and contains both crossed and uncrossed fibers (relative to side of origin). Though this relatively small tract may be important in aiding recovery of UMN function after injury to axons in the lateral corticospinal tract, it is much less helpful in clinical management and can be largely ignored. It is mentioned here only for completeness.

After a fiber is given off from the lateral corticospinal tract, it goes directly to the ipsilateral anterior horn and ends at that level, with synapses near the LMNs it will influence. Obviously, corticospinal fibers associated with neck musculature are given off first, followed in order by those associated with musculature of the upper extremity, trunk, lower extremity, and perineum. As in the lateral spinothalamic and dorsal columns systems, the corticospinal- corticobulbar fibers are somatotopically arranged. If fibers from the lateral corticospinal tract pass ventro-medially to get to the anterior (ventral) horn, where would you guess that fibers associated with the various parts of the body are located within the tract? As you can see from Figure 12, if you guessed that the fibers associated with neck are most ventro-medial, while those associated with arm, trunk, leg, and perineum are, respectively, more and more dorsolateral, you are correct! Hurray, another piece of information that doesn't need to be memorized!

In summary, the most important facts to remember about the anatomy of the corticospinal neurons are that their axons are somatotopically organized, all of them decussate at the spinomedullary junction, and they remain crossed (i.e., contralateral to their origins) until they end at various levels of the spinal cord. [Clinically, however, it is probably more practical to think of the fibers above the decussation as being contralateral, namely, to the muscles influenced, for it is the location of the muscles (effectors) on which the findings center, just as the findings in sensory systems center on the location of the receptors. Stated another way, you will reason from location of abnormal findings, largely associated with receptors and effectors (because these are where the systems are being tested during the clinical examination), back to the location of the lesion(s) responsible. If these last statements aren't clear to you, ponder them a while. Be sure to ask questions if the fog still hasn't cleared! As for function of the UMNs and LMNs, remember three things. First, both sets of neurons influence voluntary strength, muscle tone, and reflexes (among other things), as tested clinically by measuring ability to resist external force, by measuring resistance to movement when relaxed, and by measuring responses to stimulating nearby skin or stretching the muscles themselves with a reflex hammer, respectively. Second, whereas LMN findings are all basically "less than normal," some UMN findings are "greater than normal" and apparently represent disinhibition. Third, in both LMN and UMN lesions, the time-course of injuries must be taken into account to avoid being mislead by the findings.

Now, before integrating your knowledge of all three long-tract systems, pose and solve some problems related only to corticospinal and LMN fibers. A couple of starters follow:

• PROBLEM 8. A 14-year-old white male presents with paralysis of the muscles in his right leg, which came on suddenly during late summer several months ago. Now, he has no ankle jerk (L5-S1) nor patellar response (L2-L4) (DTRs) on the right, and the right foot and leg are hypotonic. He has no Babinski response on the right. The rest of the motor examination is normal. Where might his problem be located? Why?

• PROBLEM 9. A 33-year-old black female complains of progressive weakness on her left side for 3 months. The weakness initially started ir, her left fingers (C6-C8) and then spread to involve her entire left arm (C5-T1) and left leg (L1-S2). On examining her, you find paralysis of muscles in the left fingers and forearm and marked paresis in muscles of the left foot and leg. Her biceps (C5-C6), triceps (C6-C7), and supinator (C6-C7) reflexes (DTRs) are decreased on the left (O to +1), normal (+3) on the right. Her ankle jerk and patellar responses are increased on the left (+4), normal (+3) on the right. Superficial abdominal reflexes (T8-T12) are absent on the left. There is extension of the big toe in response to planter stimulation (i.e., a Babinski) on the left. There is marked atrophy of muscle mass in the left hand and forearm and occasional fasciculations in the left biceps and triceps. The rest of the motor examination was normal. Where might her problem be located? Why?

Notice further patterns regarding the long tracts. For example, (1) if you have abnormalities in all three systems and they are all on the same side of the body (below the head), though not with identical distributions in the two sensory systems, where do you think the lesion would most likely be? (Since all three systems are on the same side of the CNS from the spinomedullary junction on up, but the fibers of the two sensory systems don't integrate somatotopically till near the thalamus, the lesion would most likely be in the brainstem, on the side contralateral to the findings.) (2) If you have abnormalities in the motor system and only one sensory system, is the lesion peripheral or central? Central, of course. Why? (3) If the abnormalities (motor and in one sensory system) are on one and the same side of the body (below the head) and involve the dorsal columns system, where might the lesion be? What difference would it make if the sensory system involved were the lateral spinothalamic? Why? (4) On the other hand, if the motor abnormalities were on one side of the body and the lateral spinothalamic abnormalities were on the other side, where would you suspect the lesion to be? Why?

If you can answer the above questions handily, you should be ready to tackle specific patient problems, such as the following. I would point out that even with no further knowledge of neural structures (and their functions), integrated application of the above long tract systems will allow you to diagnose both the side of lesions and their general level--i.e., peripheral (including which spinal nerves), spinal cord (including specific level), brain stem, thalamus, or cortical (and subcortical)--in a surprisingly large proportion of patients with neurologic lesions. Even if you don't know what additional findings mean or imply anatomically, you can ignore them for now and reason based on the findings with which you are familiar.

• PROBLEM 10. A 65-year-old white female presents with sudden onset (within previous 4 hours) of inability to move left arm and leg. Upon neurologic testing, you find numerous neurologic abnormalities, including:
  1. paralysis of the left body below the neck (i.e., left hemiparesis)
  2. decreased deep tendon reflexes on the left side
  3. inability to localize light touch or pin-prick on the left side of the body
  4. paralysis of lower facial muscles on the left
  5. inability to feel light touch or sense pin-prick as sharp on the left side of the face
  6. left homonymous hemianopsia

  There was no Babinski response on either side. Where is this patient's lesion probably located? Why?

• PROBLEM 11. A 35-year-old black male presented with gradual onset over several months of weakness in his right arm and leg. His chief complaint, however, was a burn on two left fingers, which occurred yesterday while he was smoking a cigarette. On physical examination, you find marked paresis in both right limbs, decreased abdominal reflexes (superficial reflexes) on the right, increased deep tendon reflexes (biceps, triceps, supinator, patellar, and ankle) on the right, Hoffman and Babinski responses on the right. In addition, he is unable to perceive pin-prick as sharp on his left side, from his foot up to his clavicle (including his left arm), stereognosis is impossible in his right hand, position and vibratory sense are absent in his right hand and arm, and two-point discrimination and graphesthesia are impaired on his right side from the nipple line to his lower neck (including the right hand and arm). Where might his lesion be located? Why?

• PROBLEM 12. You examine a 55-year-old white male who awoke one morning, 2-3 months ago, with an inability to move the right side of his body. Admitted to the hospital at that time, he in/as found to have many neurologic findings. Presently, you find
  1. right hemiparesis of a spastic nature, but not including the facial muscles
  2. right hemianesthesia (inability to feel stimuli presented to the skin), but not including the face
  3. protruded tongue deviates to the left
  4. inability to feel pin-prick as sharp on the left face
  5. ptosis, miosis, and facial anhidrosis (Horner's syndrome) on the left

  Where might his lesion be located? Why?

We will next cover the cranial nerve systems. Don't forget to apply what you've learned from the anatomy of the long-tract systems, for many of the cranial nerve systems carry the same type of information, only localized to the head. You'll find that the anatomical patterns, too, in many ways resemble those of the analogous long-tract systems.

Even though the cranial nerves as a group do not form a functional unit, there are a few general patterns regarding their spatial organization that are helpful to know. Two will be given now. The first is perhaps so obvious that it doesn't need to be mentioned: each cranial nerve is ipsilateral to the structures it innervates. This means that if you find any abnormality, sensory or motor, on one side of the head or neck, and it's due to peripheral damage to a cranial nerve, the nerve involved is on the same side as the abnormality. This was, of course, also true of the spinal nerves discussed earlier. The second pattern reflects their rostral-caudal order of joining the CNS (Figure 13): I and II are most rostra!, joining the CNS above the brainstem; III and IV join the midbrain; V joins the mid-pons; VI, VII, and VIII join at the pontomedullary junction; and IX, X, XI, and XII join the medulla. Since they are thus approximately arranged in order. it is rather simple to recall their general location from memorizing where a few are, e.g., III-IV and VI-VII-VIII. A mnemonic (remember?) that just came to mind uses the number of syllables in a few descriptive words:

Words:	above	midbrain	pons	in between	and medulla
No. of nerves:	2	2	1	3	4

(Actually, cranial nerve XI is a bit of an oddball, for in addition to the part coming off the medulla, there's a component that comes off the C1-C5 levels of the ipsilateral spinal cord and enters the skull (through the foremen magnum) to join the fibers from the medulla. We'll discuss its anatomy a bit more later, as we consider its function.)

Since the sensory information in a given nerve (with one important exception, to be described soon) is processed at that same level or higher (i.e., more rostrally) and the motor information is also influenced primarily by structures that lie at the same level or more rostrally, abnormalities in function of that nerve can generally be assumed to reflect damage either peripherally or else centrally at or above its level of union with the CNS. Does this sound new? It shouldn't. Think about the levels of possible damage when sensory or motor abnormalities below the head were noted earlier. If the findings were referable to the distribution of certain spinal nerves, how often would the lesion be in the CNS below the corresponding segments of the spinal cord?

Let's start by considering the motor functions of the cranial nerves. It will serve as a review of motor function below the head and will also allow us to complete the "corticobulbar" portion of the corticospinal-corticobulbar (UMN) system described earlier. As you will recall, we said that corticobulbar neurons were the UMNs associated with cranial nerves, just as corticospinal neurons were the UMNs associated with spinal nerves. Also, the cell bodies of the LMNs traveling in a cranial nerve lie ipsilaterally at the level of the nerve's exit from the CNS. [One exception is the cell bodies of the LMNs in cranial nerve IV, which lie contralateral to their nerve. Fortunately, cranial nerve IV is relatively unimportant clinically . Another exception involves the cell bodies of the LMNs supplying the levator palpebrae superioris, some of which may be contralateral to their nerve (III), and those of the LMNs supplying the superior rectus, all of which may be contralateral to their nerve (III). Although these muscles are clinically important, it will turn out that other anatomical considerations make it unnecessary to remember this exception. In conclusion, then, the exceptions are rather unimportant clinically and can be disregarded. Sound reminiscent of anterior horn cells? Functionally, too, LMN lesions associated with cranial nerve function yield the characteristic findings of flaccid paralysis (Table 1) in the muscles involved. It's the UMNs--the corticobulbar fibers--that throw us a few new curves! But first, let's list the muscles we're talking about and describe briefly how to test their function. (See NEURO I-5 to learn more about clinical testing.)

As you can see from Table 2, the muscles used to test cranial nerve systems tend to group functionally according to the nerve of supply. Thus, nerves III, IV, and VI are associated with turning the eyeball, V--with moving the jaw, VII-- with making facial expressions, IX and X (and cranial portion of XI)--with swallowing and phonation (i.e., making vocal vibrations: X only), and XII--with moving the tongue. The main difference between the cortico-bulbars and corticospinals is in their level and extent of crossing the midline. Whereas most muscles supplied by spinal nerves are influenced primarily by UMNs (corticospinal fibers) originating in the contralateral cerebral cortex and decussating at the medullospinal junction; most muscles supplied by cranial nerves are influenced by UMNs (cortico-bulbars) originating in either cerebral hemisphere (roughly half of the fibers coming from each side), the fibers from the contralateral hemisphere crossing the midline at approximately the level of the LMNs they will influence. Thus, for example, corticobulbar fibers influencing the muscles of mastication come from both hemispheres, and the contralateral fibers decussate in the pons, about half-way between the midbrain and medulla. (Just so there's no confusion, remember that when we talk of corticobulbar or corticospinal fibers influencing muscles, we don't mean directly! Only LMNs innervate a muscle, so the UMNs exert their influence on the LMNs.) Thus, when most or all of the corticobulbar fibers from one cerebral hemisphere are wiped out (as happens frequently because they run in a compact bundle through the internal capsule and upper midbrain), most muscles innervated by cranial nerves may show little grossly measurable weakness, and what weakness they do show may clear within hours to a few days, presumably because of the extensive influence by corticobulbars from the other cerebral hemisphere. (For review and integration, how and why does this differ from the situation where all corticospinal fibers from one hemisphere are interrupted by a lesion above the spinal cord?)

Table 2.	Major Striated Muscle Groups Innervated by Cranial Nerves and Some of Their Functions Often Tested Clinically.
MUSCLES (OR GROUP)		FUNCTION(S) TESTED	CRANIAL NERVE
Levator palpebrae superioris		Lift upper eyelid	III
Extraocular muscles:		Turn eyeball:
	Medial rectus	Medially	III
	Superior rectus	Upward
	Inferior oblique	Upward
	Inferior rectus	Downward
	Superior oblique	Downward	IV
	Lateral rectus	Laterally	VI
Muscles of mastication:		Move jaw:
	Masseter	Upward (shut)	V
	Temporalis
	Medial pterygoid
	Lateral pterygoid	Downward (open)
	Medial & lateral pterygoids	Laterally (to opposite side)
Muscles of facial expression: e.g.,
	Orbicularis oculi	Shut eyelid	VII
	Frontalis	Wrinkle forehead
	Risorius	Smile
	Levator labii superioris	Show upper teeth
	Orbicularis oris	Purse lips
Muscles of pharynx: e.g.,
	Stylo-pharyngeus	Aid swallowing	IX, X, XI (cranial portion)
	Pharyngeal constrictors
Muscles of larynx: e.g.,
	Vocalis	Aid phonation	X, XI (cranial portion)
	Crico-thyroids
	Sternocleidomastoid	Turn head to opposite side	XI (spinal portion)
	Trapezius (upper part)	Lift shoulder
Intrinsic tongue muscles: e.g.,		Move tongue:
	Genioglossus	Forward (protrude)	XII

There are two important exceptions, where one muscle or set of muscles supplied by a cranial nerve is influenced almost entirely by corticobulbar fibers from the opposite cortex, i.e., in the pattern characteristic of muscles innervated by spinal nerves. One exception is the genioglossus, which protrudes the tongue forward and slightly to the opposite side. The other is the lower muscles of facial expression, such as the risorius and zygomaticus, which help bring the mouth corners up into a full smile, and the orbicularis oris, used to close or purse the lips.

The pattern of weakness in just the lower muscles of facial expression on one side is so important clinically that it is given a special name--central (or supranuclear) facial palsy. It is contrasted with weakness of all muscles of facial expression on one side, which is called peripheral (or infranuclear) facial palsy. The term "peripheral" facial palsy may be somewhat misleading for you, as any lesion of the facial LMNs, whether peripheral or in the brain stem, will tend to produce a "peripheral" palsy. If you remember that the central palsy is essentially an UMN problem and the peripheral palsy is a LMN problem, you should be okay.

Now let's consider a few combinations of clinical findings to illustrate the importance of the usual corticobulbar distribution and the two exceptions. In the process, we will discover two more useful rules for localizing lesions!

1. Assume that a patient's tongue protrudes to the left (abnormal: should protrude in the midline) when you ask him to stick it out. First of all, which genioglossus is weak, the left or the right? Remember, each genioglossus tends to pull the tongue forward and towards the midline. That's correct: the left genioglossus is weak. Secondly, what cranial nerve innervates the left genioglossus? It's the right XII (hypoglossal = under + tongue), isn't it? No, of course not! Remember, all structures are innervated by the ipsilateral nerves. Okay, we're concerned with the left XII nerve. Next, where does it join the CNS? In the left medulla. So at this point, with no further information, where might his lesion be? In the spinal cord? Hardly. In the left XII nerve itself? Yes that's possible. In the medulla? Yes, it could be, couldn't it. Would it be an UMN or a LMN lesion if it were in the medulla? Hard to tell. Remember, the corticobulbar fibers tend to cross the midline at about the same level that the cell bodies of the LMNs are found. So if it were a LMN lesion in the medulla, what side would it be on? The left side. What if it were an UMN lesion in the medulla? Well, probably on the right side. (Although it would be possible to hit the UMNs after they'd crossed the midline without damaging the cell bodies of the LMNs, they have a much longer course in the right medulla, before they cross.) Could the lesion be in the pons, midbrain, or cerebrum? Yes, it could. If it were at one of these higher levels, which side would it be on? The left? No, of course not. The involved UMN fibers would still be on their side of origin, so the lesion would have to be on the right. So, in summary, the lesion could be peripheral (left XII nerve, or in the muscle itself) or in the left medulla, in which case it would be an LMN lesion; or it could be in the right medulla, pons, midbrain, or cerebral hemisphere, in which case it would be an UMN lesion.

2. Now let's add another finding. Not only does your patient's tongue deviate to the left when protruded, but he is also paralyzed on the left from the neck down (i.e., left hemiparesis). Notice, I didn't specify any additional findings to help decide whether the left hemiparesis is due to LMN or UMN damage. Do I need to? Not really. Think about the distribution of damage necessary to produce hemiparesis as an LMN problem: it would have to include anterior horn cells or ventral spinal nerve roots on one side all the way up and down the spinal cord. Highly unlikely. Of course, it could be a combination of UMN and LMN, by having a left-sided lesion high in the spinal cord (perhaps C5-T1), which wiped out the lateral corticospinal tract and anterior horn cells at that level. However, that, too, is much less likely than a lesion above the spinal cord. So, without specifying anything further, the finding of hemiparesis nearly always reflects UMN damage. [By the way, this may be a good time to become familiar with the family of paresis/plegia terms. Plegia is used as synonymous with paresis--incomplete paralysis--, although plege = stroke, so describes a common cause rather than the problem itself. Hemi = half, and hemiparesis or hemiplegia denotes paresis of one side of the body. Notice, however, that it doesn't include the head; for practical purposes, it can be thought of as paresis of both extremities on one side of the body. Paraplegia refers to paresis of both lower extremities, quadriplegia (quadri = four) refers to paresis of all four extremities, and monoplegia (mono = one) refers to paresis of only one extremity.]

   Assuming that there is only one lesion and that it's all on one side of the nervous system, where might it be located? Can we still entertain the possibility of a peripheral lesion, in left XII? No. How about in the left medulla? Why not? Because the corticospinal fibers in the left medulla would decussate at the medullospinal junction, influencing muscles on the right side of the body. Is the right medulla a possibility? Yes, it is. Remember, above the level where the UMNs associated with the genioglossus cross, those corticobulbar fibers are on the same side as the corticospinal fibers that will decussate to influence muscles on the left below the head. Remember, also, that the corticospinal-corticobulbar fibers run together in somatotopic order. So a rather small lesion could wipe out all of the UMN fibers in the right medulla and produce the findings shown by our patient. How about higher levels? Yes, the lesion could be in the right pons, midbrain, or cerebral hemisphere, for the UMN fibers influencing the left genioglossus and left body below the neck remain on the right all the way to their origins in the cortex.

3. What if the patient's hemiplegia had been on the right instead of the left? We still know that we're dealing with a CNS problem above the spinal cord, right? (If you're not sure, take time out to review as necessary to convince yourself of that much. If you still aren't convinced after careful review, be sure to work it out with another student or an instructor.) Could the lesion be in the left medulla? Yes, it could. Why? Well, the corticospinal fibers in the left medulla will decussate to influence muscles on the right, so damage to them could produce a right hemiplegia. At the same time, LMNs that will travel in the left nerve XII to the left genioglossus originate in the left medulla. How about the right medulla? That doesn't work, does it, for the corticospinal fibers in the right medulla influence muscles on the left side of the body. As for higher up on the right side, the same argument holds: corticospinal fibers there influence muscles below the head on the left side of the body. How about the left pons, midbrain, or cerebral hemisphere? The right hemiplegia would be consistent with a lesion there (why?), but what about the weak left genioglossus? No, LMNs traveling in a cranial nerve originate only at the level of the nerve--in this case, in the medulla.

   Notice what a striking difference it makes, whether the hemiplegia is on the same side as the weak genioglossus or on the opposite side. It turns out that this reflects one very useful pattern. Whenever a motor problem referable to a cranial nerve (you will find out later that the same rule holds for sensory problems) is combined with hemiparesis on the opposite side (so- called crossed paresis), the lesion is at the level of the involved cranial nerve and on the same side as the cranial problem. This is because the LMNs traveling in the nerve are found only ipsilaterally and at that level, while the corticospinal fibers on that same side have not yet decussated. Since all LMNs traveling in cranial nerves arise only in the brainstem (with the exception of the spinal component of XI), crossed paresis is a hallmark of a brainstem lesion. To make sure you understand this important point, work out the location of the lesion in a patient with:

   1. right hemiparesis and weakness of all muscles of facial expression on the left.

   2. left hemiparesis and weakness of muscles of mastication on the right.

   3. left hemiparesis and paralysis of upward and medial gaze in the right eye.

   4. right hemiparesis and absence of gag reflex (cranial nerves IX and X) on the left.

   Got it? Good! Let's go on.

   The converse of the above principle is also important. That is, whenever a motor problem referable to a cranial nerve is combined with hemiparesis on the same side, the lesion is above the level of the involved cranial nerve and on the side opposite the motor findings. Why? Because to be on the same side as the hemiplegia, which is caused by damage to corticospinal fibers on the opposite side, the cranial nerve problem must also be due to UMN damage, occurring before the UMNs associated with the cranial nerve have crossed the midline. This is where it is important to remember the bilateral UMN supply to most cranial LMNs. For example, if there were a lesion in the left upper medulla that wiped out the left cortico-spinal-corticobulbar fibers, would you see motor problems referable to cranial nerves IX and X (e.g., absent gag reflex), XI (e.g., inability to turn the head forcibly to the side), and XII (e.g., inability to protrude the tongue in the midline), along with the right hemiparesis? For nerves IX, X, and XI, the answer is "probably not." LMNs coursing in these nerves on the right receive considerable input from the right corticobulbar fibers (uncrossed), and LMNs coursing in these nerves on the left also receive considerable input from the right corticobulbar fibers (crossed). Only the LMNs supplying the genioglossus on the right (in nerve XII) have very little UMN input, because that input is almost entirely crossed--i.e., from the left corticobulbars. Similarly, if there were a lesion in the left upper pons that wiped out the left corticospinal-corticobulbar fibers, would you see motor problems referable to cranial nerves V (e.g., can't bite down forcibly), VI (e.g., can't gaze laterally), and VII (e.g., can't shut eyelids forcibly, give a big smile, or show his upper teeth well), as well as to cranial nerves IX-XII? Again, the answer for nerves V, VI, as well as IX-XI, is "probably not," for the same reason given above. In addition to the right hemiplegia and deviation of the protruded tongue to the right, only the muscles of facial expression in the lower right face would be weak, again because UMN input to their LMNs (in nerve VII) is almost entirely crossed. [It is probably not too soon to alert you that the neural regulation of the extraocular muscles is somewhat more complex, once you get above the LMN level. A set of neurons in the brainstem link together the actions of the various extraocular muscles, so that they work as a group and the movements of the two eyes are coordinated. As a result, any UMN lesion that does produce a deviation of normal gaze (and there are some, as you'll find out later), tends to produce a "conjugate" deviation. In other words, the eyes still move together, so as to maintain binocular fixation on objects. For example, if you find that a patient can't turn his left eyeball laterally, although his right eyeball turns medially (i.e., normally) as he attempts to look to his left, his gaze is disconjugate, suggesting an ipsilateral LMN problem--in this case, of the LMNs coursing in the left nerve VI. If, on the other hand, when the patient tries to look to the left, neither eyeball turns (i.e., medially for the right, laterally for the left), he has a conjugate gaze paralysis--connoting UMN damage. We'll discuss movements of the eyes more later. For now, simply realize that conjugate gaze paralyses reflect possible damage at various levels, from the pons on up to the cerebrum, and concentrate instead on using crossed paralyses and unilateral combinations of hemiplegia, genioglossal weakness, and central facial palsy (i.e., of lower face only) to reason out the locations of lesions.]

   If we go even higher, wiping out all of the corticospinal-corticobulbar fibers in the left midbrain or left cerebral hemisphere, would we find any additional motor problems referable to cranial nerves? No, because the LMNs in cranial nerves III and IV on either side also receive considerable bilateral UMN input (though not directly: to be discussed later). Consequently, if a patient has hemiplegia, genioglossal weakness, and paralysis of the lower facial muscles-- all on the same side--, we know the lesion is at least as high as the pons and is on the side opposite the findings, but we can't locate it more precisely-- e.g., in the pons, midbrain, internal capsule, corona radiate, or cerebral cortex--without additional information.

4. Let's go back to the patient with a right hemiparesis and tongue deviating to the left upon protrusion. Right away, we can argue for a left-sided medullary lesion, as the patient shows a classic crossed paralysis. Fine, but now let's ask ourselves some additional questions. What if instead of a complete right hemiparesis, the patient had had only a partial--monoplegia of the right arm, for example? Or what if instead of motor findings on the right, there had been right sensory losses--e.g., right analgesia or right hemianesthesia? Here, another very useful pattern surfaces. No matter what the problem below the head, whether involving the lateral spinothalamic system, the dorsal columns system, or the motor system, presence of a cranial motor problem on the opposite side, i.e., a motor problem referable to a cranial nerve on the opposite side (you will find out later that this pattern holds for sensory findings referable to cranial nerves, also) indicates that the lesion is at the level of the cranial nerve involved and on the same side as the cranial nerve problem.

   This pattern is essentially an extension of the one discussed above in 3. Let's reason it out for our patient with right hemiparesis and tongue deviating to the left upon protrusion. If the paresis involved only the right arm, the tongue problem would still indicate a lesion above the spinal cord. At that level, the arm weakness would have to reflect an UMN problem, so points to the left side of the CNS--partial destruction of the somatotopically arranged fibers in the left corticospinal tract. Since the tongue problem should also reflect damage on the left, it is probably an LMN problem--thus, located at the level of the involved cranial nerve (XII) in the medulla. If we now take away the patient's hemiparesis and give him right-sided sensory abnormalities below the head, we find that our reasoning remains basically the same. Since the tongue problem still indicates a lesion above the spinal cord and both the lateral spinothalamic and dorsal columns systems have decussated by this time, thus lying on the same side as the left corticospinal tract we discussed above, the sensory problems also reflect damage to the left side of the CNS, consistent with a LMN weakness of the left genioglossus.

   To generalize and state the pattern another way, since the ipsilateral dorsal columns system, lateral spinothalamic system, and corticospinal fibers above the spinal cord all influence the same (contralateral) side of the body below the head, findings involving them on one side of the body below the head become almost equivalent to each other when reasoning out locations of CNS lesions known to be above the spinal cord. Again, the converse is also true. Remember that hemiparesis and cranial motor findings on the same side suggest a contralateral lesion above the level of the involved cranial nerve. If we now substitute an ipsilateral monoplegia or ipsilateral sensory abnormalities below the head for the hemiparesis, the cranial motor findings still suggest a contralateral lesion above the level of the involved cranial nerve. Be sure that all of the preceding makes sense to you before going on. Explaining it to a colleague is a good way to test your ability to use the information.

First, let's consider the similarities between the trigeminal sensory system and the two sensory long-tract systems, because they will serve as a partial review and point up information that does not have to be learned separately for the trigeminal sensory system. The cell bodies of the first-order neurons are unipolar and reside in a sensory ganglion (called the trigeminal, semi-lunar, or Gasserian ganglion) located near the CNS. (The exception to this, namely, cell bodies of first-order proprioceptive fibers from muscles innervated by V, were mentioned earlier. They are rather unimportant clinically, but do reside inside the CNS, in case you want to know.) The first-order fibers pass through the sensory ganglion and enter the CNS (mid-pons level, remember?) to synapse on second-order neurons ipsilaterally. The axons of the second-order neurons soon decussate as they ascend toward their relay nucleus in the thalamus--the ventral posteromedial nucleus (VPM). The VPM is just medial to the VPL--the thalamic relay nucleus in the lateral spinothalamic and dorsal columns systems. In fact, the paired VPM and VPL nuclei on either side are functionally a single somatotopically-organized nucleus. In the VPM lie the cell bodies of third-order neurons, whose axons ascend in the somatosensory radiations (first through internal capsule, then corona radiate) to the ipsilateral post-central gyrus (but contralateral to the first-order neurons!) of the parietal lobe of the cerebral cortex, maintaining somatotopic relations with somatosensory fibers from other parts of the contralateral body all the way. There's even one further anatomic similarity to the sensory long-tract systems: trigeminal first-order axons entering the CNS split up to begin two anatomically distinct subsystems, roughly analogous to the lateral spinothalamic and dorsal columns systems, respectively, which go their separate ways and do not reintegrate somatotopically until they near their thalamic relay nucleus.

Before we trace the two distinct subsystems from entry into CNS up to contralateral VPM nucleus, let's give their names and discuss their functions briefly, for here we will find one important difference from the analogous sensory long-tract systems. One subsystem is called the principal (or chief) sensory subsystem, it carries the sensory modalities handled by the dorsal columns system--namely, vibratory sense, position sense, and discriminative touch. The other is called the descending (or spinal) subsystem; it carries the sensory modalities handled by the lateral spinothalamic system--namely, pain and temperature--and anterior spinothalamic system--namely, crude touch. The important functional difference from the analogous long-tract systems is as follows. Whereas a lesion of the dorsal columns system by itself usually does not produce a deficit in simple touch, as might be tested with a wisp of cotton, a lesion of the principal trigeminal subsystem does produce such a deficit. Apparently, the information on simple touch carried in the descending trigeminal subsystem can not substitute completely for similar information carried in the principal subsystem. For this reason, presumably, the routine clinical test for the principal trigeminal subsystem simply involves touching various areas of the face systematically with a light stimulus (e.g., a wisp of cotton) and comparing the intensities of sensation reported from the different areas. No tests for vibratory sense, position sense, or discriminative touch are routinely done.

The essential anatomical difference between the two subsystems can be expressed in terms of the levels at which anatomical cell bodies of the second-order neurons lie. Those for the principal subsystem (in the principal sensory nucleus) lie at the level where the trigeminal nerve joins the CNS--i.e., mid- pons, whereas those for the descending subsystem (in the descending nucleus of the trigeminal) lie strung out from the lower pons all the way down to the first two cervical segments of the spinal cord. As a consequence, the first-order axons in the descending subsystem do just that, descend to various levels in lower pons, medulla, or cervical cord, to reach the second-order neurons. As mentioned earlier, the axons of the second-order neurons in both subsystems tend to decussate just shortly above (rostra! to) the c