At the request of several members, I am writing a summary of conus medullaris injuries, its clinical manifestations and what potential therapies are being developed to restore function in such injuries.
Most spinal cord injuries involve the cervical or mid-thoracic spinal cord. However, about a third of spinal cord injuries affect the lower thoracic spinal segments where the lumbosacral spinal cord resides. The lumbosacral spinal cord starts at T11 and continues to L1 and part of L2, where the end of the spinal cord called the conus medullaris is located. The conus medullaris contains the sacral segments (S1-S5). If the injury involves T12, there may be loss motor and sensory function in the lumbar segments. If the injury involves L2, cauda equina injuries may be associated.
L1 spinal segment injuries cause a stereotyped neurological syndrome composed of three signs. First, there is a ?saddle? shaped sensory loss, called this because the person loses sensations from parts of the body that contact a saddle, when sitting on the saddle of a horse, i.e. the perineal region and the inner butt. Second, movement of the proximal leg muscles (hip flexors, quadriceps) are usually preserved while distal muscles (ankle) may be affected. Third, bladder and anal functions are frequently severely affected, usually with frank incontinence of bladder and bowel.
Mild injuries of the conus may be subtle, i.e. changes of sensation in the perineal region. This may manifest as changes in the way toilet paper feels when wiping or leaking of urine when the bladder is full, or difficulties starting or stopping urine stream. Physical examination may reveal diminished ankle and plantar reflexes, altered sensations in lateral foot (S1), weakness of voluntary toe movement, and numbness of the penis or vagina. Loss of erection is frequent. Urinary retention with overflow incontinence, as well as fecal incontinence usually develops. Pain is usually in the lower back.
An injury of the cauda equina can be differentiated from a conus medullaris syndrome by several key differences. First, manifestations of cauda equina injuries tend to be assymetrical, i.e. affect one side more than others. Conus medullaris syndrome is almost always affects both sides equally. Second, muscle atrophy affecting one side more than the other are common. Third, knee and ankle jerks are both affected in cauda equina injuries whereas knee jerks should not be affected by conus injuries. Likewise, more proximal muscles may be affected in cauda equina. Cauda equina injuries are more likely to be associated with severe radicular pain.
Acute treatment depends on the cause. If a mechanical cause is suspected, lateral x-rays and MRI scans should provide evidence of spinal stenosis, bony fracture with compression, herniated disc at L1, tumor compression of the conus, spinal cord injection or abscess, spina bifida, hemorrhage, lipoma, arteriovenous malformations, sarcoidosis, and deep venous thrombosis of spinal veins. However, the MRI may not reveal any cause of conus injury in about half of cases. Obviously, if there is something compressing the conus, it should be removed and appropriate therapies applied to prevent recurrence.
Therapy for chronic conus medullaris currently consists of teaching patients how to deal with the consequences of the injury, i.e. self-catheterization, manual fecal dis-impaction, prevention of decubiti, strength-training of the remaining muscles, and steps to address erectile dysfunction. At the present, there is no therapy aimed at restoring function after conus medullaris injuries. In part, this is because treatment will require neuronal replacement, as well as axonal regeneration. Effective treatments for both of these have not yet been shown in human clinical trials. Relatively few studies are being carried out even in animal models.
My laboratory has consequently focused on conus and cauda equina injuries in the past three years. Our first goal is to establish lumbosacral spinal cord injury model and the behavioral consequences in rats. To do this, we dropped a 10 gram weight onto the T13-L1 junction of the rat spinal cord. Note that rats have a T13 segment while T12 segment is usually the lowest segment in humans. Hitting the spinal cord in the T13-L1 junction damages the L4-S5 spinal cord in rats, causing loss of neurons as well as sensory and motor fibers coursing through the region. The rats often develop flaccid bladders. They can walk but abnormally with weak anterior tibialis, gastrocnemius, and toe muscles. These can be readily quantified by footprint analysis.
Regeneration alone is unlikely to cure conus medullaris injuries. Regenerating fibers have no target to connect to. Neuronal replacement and rebuilding the neuronal circuitry of the lower spinal cord will be necessary to restore function. In addition, we must induce the replaced neurons to send axons to connect to the appropriate neurons and muscle. Finally, there is the issue of atrophied muscles and whether they have appropriate sites for regrowing motoneuronal axons to connect to. For example, peripheral nerve surgeons have long been convinced that motor recovery more than 3 months after a peripheral nerve reconnection is unlikely because the muscles have undergone atrophy and no longer have the motor endplates for regenerating motor axons to synapse on.
We are currently replacing motoneurons in the lumbosacral spinal cord by transplanting neural stem cells and using a variety of treatments to stimulate the cells to extend axons out of the ventral roots to reconnect to muscle, as well as regeneration of ascending and descending fibers. Our initial studies using neural stem cells obtained from the subventricular zone but we are also attempting to create neurons from Muse (multipotent stress enduring) cells that can be isolated from bone marrow, adipose tissues, and umbilical cord blood. Finally, we are looking the possibility of using Muse cells to restore atrophied muscles.
The experiments are still in their early phases and we don?t have treatment effects to report yet. However, we are very excited by some early successes to date. First, we have successfully developed ways to grow Muse cells from bone marrow and umbilical cord blood. Second, Dezawa, et al. who discovered Muse cells have results suggesting the Muse cells are effective for stroke and muscle replacement. Third, we are finding strong evidence of walking, bladder, and bowel recovery in patients treated with umbilical cord blood mononuclear cells. So, I am personally very hopeful that our experiments will yield interesting and possibly promising results.
What is the likelihood of effective therapies in the coming years? I believe that neuronal replacement therapies will be available in the near future. The obstacles were immune-compatible sources of the cells, inability to get the neurons to grow their axons to connect with the proper targets, and the lack of good animal models to assess the therapies. The development of iPS cells, adult pluripotent stem cells such as Muse cells, and availability of HLA-matchable sources such as umbilical cord blood will overcome the cell source obstacle. Methods of stimulating cells to grow are available, including lithium and PTEN. Animal models are being developed.