More active assisted movements are intensively studied for potential benefit on motor control, not just spasticity per se. This includes a range of therapist-guided and robotic-assisted options for functional upper and lower extremity movements (see chapters 22 and 23). For example, body weight-supported treadmill training is directed at lumbar locomotor central pattern generator (CPG) circuits, with the goal of improving CPG coordination and ambulation [76–82]. With more extensive experience in the stroke population, robotic-assisted hand and arm therapy is directed at improving functional reach and grasp motions [83–87]. Both types of therapy may reduce spasticity by replacing aberrant hyperexcitable sensory circuit firing with more functionally appropriate afferent signaling, leading to less muscle spasm and co-contraction [88, 89]. These assisted, pattern-based interventions are thought to induce beneficial neuroplasticity through task-specific learning [41, 90]. It should be noted that to date, the benefits of activity-based therapy have been greater for those individuals with clinically motor incomplete SCI [91, 92]. However, optimism should be retained for those with clinically complete SCI, as combinations of physically assisted therapy with drugs and electromagnetic stimulation may synergize to improve movement in the future even after severe SCI.

Pharmacological approaches to treating spasticity are regularly reviewed [96–98], so instead this chapter will discuss these agents from the perspective of their impact on the broader issue of motor control after SCI. Compared to physical and other non-pharmacological approaches, drug studies are much easier to standardize in terms of dosing, scheduling, and administration; pharmaceutical companies offer an enormous amount of financial and human capital to back drug studies that cannot be matched by smaller device firms and granting agencies to back non-pharmacological studies; and once at steady state, drugs can exert their influence all 168 h of every week, rather than the handful of hours (at most) that are usually devoted to nondrug interventions. On the other hand, drugs have disadvantages that include more potential for side effects and drug-drug interactions, never mind the general neural inhibition mediated by many spasticity drugs (sedating effects).

The tissues targeted by different anti-spasticity medications vary. While it is not the muscles, the neuromuscular junctions, nor the peripheral nerves that are the site of pathology in SCI, they are often selected for pharmaceutical intervention, in part because they are more accessible than the CNS, and the functional effect can be similar. In some ways, however, this is adding insult to injury, further weakening a damaged nervous system rather than addressing the problem at the site of pathology. Dantrolene, botulinum toxin, and phenol injections weaken muscles, block neuromuscular junctions, and preclude peripheral nerve signal conduction, respectively, and thereby weaken spastic muscles. While decreasing the effects of abnormal motor output, however, they also diminish the capacity for more physiologically appropriate motor production and thereby theoretically limit recovery. We feel these are best used either as add-on medications when others directed at CNS circuitry have failed or when botulinum toxin injections are targeted at a specific, overly active muscle or sub-portion of a muscle that is reliably perturbing a specific motor function (discovered perhaps through poly-electromyography applied during specific tasks).

Pharmacological agents that address CNS neural circuitry are more directed toward reimposing inhibition at the site of pathophysiology in SCI. Nevertheless, the use of these medications may resemble using a sledgehammer as a fly swatter, as none of the medications are precise enough. It is nearly impossible to use these agents to impact just the altered neural circuitry leading to hyperreflexia or hypertonia or disordered coordination between motoneuron pools (dyssynergias) without also decreasing physiologically appropriate motor output. Whether we are talking about pre- or postsynaptic inhibition or action at GABA A, GABA B, alpha, 5HT, or dopamine receptors, drugs like benzodiazepines, baclofen, tizanidine, cyproheptadine, or dopamine agonists (respectively) should be used at minimally effective doses for the individual patient to best improve functional status. There is clinical experience, if not randomized clinical trials, to suggest that combinations of these drugs may be as or more effective than high doses of just a single line of pharmacological attack.

To better deliver drug to the site of impaired neural circuitry, baclofen can be given intrathecally instead of orally [99]. Intrathecal delivery trials can be pursued to ascertain clinical effect prior to the permanent implantation of an intrathecal pump, but it seems this course of action is often pursued before a full exploration of the effect of oral medication combination therapy. While intrathecal pumps can be very effective, and appeal to many with difficult to control spasticity, there is a rare but real risk of life-threatening baclofen withdrawal syndrome due to pump failure, disconnection, or poor management. All should be aware that pruritus in the absence of a rash in an intrathecal baclofen user is withdrawal until proven otherwise and that oral replacement is often insufficient in these cases. If it is not feasible to reestablish intrathecal delivery quickly (risk of infection with pump pocket abscess, for instance), intravenous benzodiazepine therapy may be necessary to stabilize the baclofen withdrawal patient.

One advantage to intrathecal delivery of baclofen is that it can be combined with intrathecal delivery of medications for neuropathic pain such as opioids, clonidine, and local anesthetics (“-caines”) [100, 101]. This raises the issue of the interaction between pain and spasticity. Many patients use changes in their spasticity as an indication of other noxious stimuli or pathology they cannot consciously feel. Certainly, both muscle afferents and cutaneous nociceptive afferents can activate involuntary motor output in SCI, so conjunctively treating pain and spasticity in parallel makes sense. In patients taking gabapentin or pregabalin for chronic pain, they often report improvements in spasticity. When necessary, a urinary tract infection associated with worsened spasms should be treated not only with antibiotics but also with a short course of increased pain medication along with a short-course increase in anti-spasticity medication dosing.

Some electromagnetic stimulation paradigms are hailed as panaceas for pain, spasticity, and motor skills, but as with most non-pharmacological interventions, the evidence lags the hype. This is partly due to the infinite number of stimulus parameters that may be applied, leading to difficulty in comparing results from numerous small studies using different paradigms. We will briefly touch on several popular (or promising) electromagnetic interventions:

At a more fundamental level though, by targeting peripheral nerves, FES bypasses all of the intrinsic circuitry that the central nervous system has evolved to carry out movements in the most efficient and coordinated manner. Successful modeling has not yet been achieved to recapitulate all of the feedback and feedforward loops that regulate groups of synergistic agonists and antagonists, that stabilize proximal muscle groups while distal muscles execute fine tasks, and that adjust movements in response to incoming sensory information. Whether human technology can outdo millions of years of evolution within the next several decades remains to be seen.

Both invasive and noninvasive spinal cord stimulation appear to activate local afferent circuits that trans-synaptically amplify signals transmitted along spared descending fibers. The cited studies have applied stimulation at various sites along the cord, in patterns ranging from constant direct current to pulsatile frequencies between 10 and 1500 Hz. When targeted caudally to a lesion to combat lower extremity spasticity, frequencies of 50–100 Hz applied to the upper lumbar region are most effective, whereas frequencies of 20–60 Hz most effectively engage the locomotor CPG [139].

It remains to be seen whether a “sweet spot” can be found in which stimulation simultaneously reduces spasticity and enhances volitional motor control over multiple muscles or whether stimulation patterns will need to be dynamically adjusted depending on the task being performed. However, its ability to facilitate proper functioning of excitatory and inhibitory circuits simultaneously in their naturally intertwined milieu imparts spinal cord stimulation with extraordinary potential to improve all aspects of motor control after CNS injury.