This multicenter study confirms the previous evidences of the effectiveness of nabiximols in reducing spasticity in MS patients. We found that 77% of patients reached the IR threshold and 22% the CRR threshold at the end of the 4 weeks trial period. The mean NRS spasticity score reduction was 23.7% at T1 and a 27.6% at T2 (3 months) compared to T0 (baseline). These results were in line with a larger multicenter observational study on the use of nabiximols in Italy [11]. In this latter study, 70.5% of patients with similar baseline characteristic reached IR threshold. Discontinuation of nabiximols during the observation period proved also to be on the same level of the Italian Study (36% versus 39.5%).

The most common adverse events (AEs) were mild to moderate dizziness and fatigue, with overall AEs reported by 47% of patients. Recently, observational studies have shown that patients tend to use lower doses of nabiximols compared with clinical trials, resulting in a lower incidence of AEs, with no evidence of addiction, abuse or misuse [10].

Patients undergoing PT intervention combined with nabiximols had higher probability of having a CRR, compared to those treated only with nabiximols. A recent meta-analysis demonstrated that PT interventions are a safe and beneficial option for spasticity in patients with MS [14]. Due to the multicenter nature of our survey, we considered a heterogeneous array of PT programs including all our patients; accordingly, we preferred to divide bimodally the study population in those attending or not attending PT. PT interventions included exercise programs on self-perceived spasticity and muscle tone respectively in both outpatient and home-based regimens. We have previously demonstrated that 15 PT sessions in addition to botulinum toxin type A injection were more effective than botulinum toxin alone, suggesting a potential synergistic effect [19].

A possible explanation of the synergistic effect of oromucosal cannabis with rehabilitation treatment was suggested by Mori et al. [20]. Exercise increases endocannabinoid signaling in humans [21] and attenuates clinical manifestations of experimental MS [22] through increased Cannabinoid receptor type 1 (CB1) sensitivity [23]. In contrast, genetic deletion of CB1 receptor is associated with decreased improvement of motor performance induced by exercise [24]. Moreover, the CB1 receptor seems to be involved in the control of spasticity both in EAE [25] and in patients with MS [11, 24].

The probability of being late discontinuers was reduced in patients undergoing combined therapy compared to those treated with nabiximols alone confirming that a combined treatment may increase effectiveness on MS spasticity. The major cause of late discontinuation in our sample was, indeed, related to the lack of effectiveness. These results are in line with our previous observations showing that the major determinant of botulinum toxin type A discontinuation in clinical practice was the lack of regular PT [26].

The retrospective design of our study implies a number of limitations. The main drawback regards the absence of a standardized and monitored PT program followed by all patients. Each PT program was selected and customized accordingly to the participant's specific needs and physical capabilities but included a minimal set of activities (see Design and Setting). However, the complete list of PT followed by our group of patients cannot be obtained: a prospective assessment aimed to evaluate specific techniques’ single contribution to the PT improvement of spasticity in MS patients treated with anti-spasticity drugs would be required to properly assess this issue.

The advantage of this design is to reflect more closely the real clinical setting of rehabilitation in MS patients, but future randomized, controlled trials are required to explore the impact of pre-defined specific PT programs applied to MS patients treated with Nabiximols.