The pathophysiology of spasticity has been reviewed recently in a comprehensive two-part article by Gracies (2005a, b). In brief, damage to the central nervous system leads to acute and chronic changes. The acute effects include paresis and short-term immobilization, whereas chronic effects include plastic rearrangements in the CNS as the result of either CNS injury and/or chronic disuse (Fig. 11.1). These changes influence the innervation of the muscles and the reflex arch leading to spasticity, spastic dystonia, and spastic co-contractions. The end result is muscle shortening and contracture caused by chronic spasticity and muscle disuse.

The exact mechanisms through which a state of muscle hyperactivity and spasticity develops after CNS injury are still unclear. As emphasized by Gracies (2005b), extensive sprouting and new synapse formation may play an important role in inducing overactive stretch reflex since the new connections are often hyperexcitable and may act differently from those lost secondary to CNS damage (Gioux and Petit 1993). There is some evidence for both decreased reciprocal I a inhibition (which inhibits alpha motor neurons via a disynaptic interneuron) and decreased I b, nonreciprocal inhibition (which via activity of Golgi tendons limits limb extension), suggesting contributions from these mechanisms to the increased stretch reflexes in spasticity (Crone et al. 2003). Furthermore, muscle immobility (as seen in spastic paresis) increases the discharge of muscle spindles (Williams 1980) which via the gamma system can lead to increased stretch reflexes and increased muscle tone. Finally, electrophysiological studies of patients with spastic hemiplegia indicate hyperexcitability of small group II afferents (originating from spindle’s secondary endings) which in a normal state inhibit motor neurons via spinal interneurons (Marque et al. 2001). The function of these type II afferents is modulated and inhibited by descending rubro- and vestibulospinal pathways that often get damaged in CNS injuries.

On the other hand, Renshaw cell inhibition (RCI) and direct alpha motor neuron hyperexcitability do not seem to play a major role in spasticity. In fact, in human, RCI has been shown to increase after CNS damage and in the presence of spasticity (Katz and Pierrot-Deseilligny 1982).

Spasticity may cause pain through a variety of mechanisms. Some spasticity-related pain (SRP) occurs in the form of muscle spasms caused by increased muscle tone and enhanced reflex activity. The pain could arise from the affected joints that are limited in movement by the attached stiff, spastic muscles. The frequent pain from spastic muscles and painful joints can also set in motion spinal and supraspinal circuits which cause central sensitization leading to pain chronicity (Chap.​ 2). In small children, adductor spasticity could lead to hip subluxation and pain.

Treatment of spasticity is heavily weighed on pharmacological agents which can cause muscle relaxation. The commonly used drugs for treatment of spasticity include GABAergic agents such as baclofen and benzodiazepines. Tizanidine, an alpha adrenergic drug and a potent muscle relaxant, is also widely used. Unfortunately, severe spasticity often requires larger doses of these medications that are beset by emergence of undesirable side effects (sedation, hypotension). Severe and advanced cases of spasticity (especially in the lower limb) may require baclofen pump placement. Although treatment of spasticity may alleviate the associated pain, in most cases, addition of analgesic drugs is required. The commonly used pharmacological agents include tricyclic antidepressants, nonsteroidal anti-inflammatory agents, and, in severe cases, opioid analgesics.

This section covers the blinded studies that assessed pain in adults’ upper and lower limb spasticity and in spasticity-related pain of children with cerebral palsy.

A total of nine blinded and controlled studies included pain assessment among the assessed variables evaluated in the investigation and reported in the results. All nine studies found BoNTs (A and B) to be effective against upper limb spasticity (Brown et al. 2014) which led to the FDA approval of onaA for treatment of upper limb spasticity.

Bakheit et al. (2001), in a blinded, controlled study of 59 patients, first reported on evaluation of BoNT efficacy against pain associated with spasticity. The pain was assessed on a 0–3 scale (no pain to severe pain). AbobotulinumtoxinA, 1,000 units, was injected into different arm and forearm muscles. The authors noted no improvement of pain after aboA administration. In 2004, Childers et al. and Brashear et al. also found no pain improvement in their studies of 91 and 15 patients, respectively. The former authors used three different doses of onaA (90, 180, and 360 units), while the later employed two doses of rimaB (5,000 and 1,000 units). Childers et al. (2004) assessed pain through a 0–4 scale with four being severe pain, whereas the exact method of pain assessment was not defined in Brashear et al.’s study (2004). In agreement with the above studies, another more recent study which assessed the efficacy of aboA (1,000 units) in 55 patients with spasticity also failed to note any significant improvement in pain after administration of the BoNT into the muscles of the upper extremity (Lam et al. 2012). Since the study was conducted in noncommunicative patients, the evaluation of pain was conducted via an observational 0–5 scale (PAINAD).

In contrast to the aforementioned studies, five other blinded studies of BoNTs and spasticity, four using aboA and one rimaB, reported significant improvement of pain after administration of BoNTs into the upper limb muscles. Four of five of these studies used the validated and widely used visual analog scale (VAS).

Suputtitada and Suwanwela (2005), in a study of 50 patients, reported significant improvement of pain after administration of aboA (three doses: 375, 500, and 1,000 units) injected into spastic upper limb muscles. The positive result of this study was supported by another blinded study (Marco et al. 2007) of aboA in spasticity that used 500 units. The assessment tool for pain was VAS in both studies.

Another two blinded studies of BoNT treatment in spasticity that assessed pain via VAS also reported significant pain relief. Shaw et al. (2011) enrolled 333 patients with spasticity in a prospective, placebo-controlled, blinded study. Patients received either placebo, 100, or 200 units of aboA in the spastic upper limb. Reinjections were performed at 3, 6, and 9 months. A significant improvement of pain was noted at 12 months but not at 1 and 3 months. In another blinded study of 163 patients (Rosales et al. 2012), administration of 500 units of aboA into the arm and forearm muscles caused significant pain relief at 4 and 24 months. Marciniak et al. (2012) assessed pain through the short-form McGill Pain Questionnaire in 37 patients with post-stroke shoulder spasticity who participated in a double-blind trial investigating the efficacy of onaA (140–200 units) in spasticity. At 4 weeks, pain was significantly reduced (P < 0.05) compared to baseline, but the placebo group also demonstrated the same degree of pain reduction.

A recent double-blind study (Gracies et al. 2014) that used rimaB toxin (5,000 and 10,000 units) in elbow flexors also demonstrated significant reduction of pain at 1 month following toxin injection (P = 0.017). The main features of the nine double-blinded BoNT spasticity studies that have reported on the results of pain assessment are presented in Table 11.1.

At first glance, the results of BoNT treatment in spasticity-related pain from the nine blinded studies mentioned above may appear controversial or contradictory (four against and five in favor). A more careful evaluation of these studies, however, provides useful explanations for the apparent contradictory results. All four studies that used an established and validated pain assessment tool (in this case VAS) reported efficacy against pain. Using the assessment criteria of the American Academy of Neurology (Appendices 3.​1 and 3.​2), the level of evidence for efficacy for aboA in spasticity-related pain of the upper limb (using VAS for pain assessment) is A (established efficacy based on two or more class I studies). The delayed efficacy (at 12 months, probably after third injection) in the study of Shaw et al. (2011) is most probably related to the small dose of aboA used by the investigators (100 and 200 units versus 500 and 1,000 units used by others). Such cumulative, late effect after repeat injections has been reported in other pain indications after BoNT treatment especially with onaA administration in chronic migraine (Aurora et al. 2014). The efficacy of aboA in relieving spasticity-related pain is supported by a recent large prospective, open-label European study (Jost et al. 2014) of 408 patients in which 58.9 % of the patients reported pain relief. Evaluation of the efficacy of the other forms of botulinum neurotoxin in spasticity-related pain deserves further investigation via placebo-controlled studies.