Pisa Syndrome (PS) is a lateral trunk flexion frequently associated to Parkinson's disease (PD). The management of PS is still a challenge for the clinician, because it poorly responds to anti-parkinsonian drugs, and the improvement achieved with neurorehabilitation or botulinum toxin injections tends to fade in 6 months or less. Transcranial direct current stimulation (t-DCS) is a non-invasive neuromodulation technique, with promising results in movement disorders. Aim of our study is to evaluate the role of bi-hemispheric t-DCS as add-on to neurorehabilitation in PS. Twenty-eight patients affected by PD and PS were managed with a 4-week hospital neurorehabilitation programme and randomized to: 1) t-DCS group: 5 daily sessions (20 minutes - 2 mA) with cathode over the primary motor cortex (M1) contralateral to PS, and anode over the M1 cortex ipsilateral to PS; or 2) sham group. Patients were tested with kinematic analysis of trunk movement in static and dynamic conditions, UPDRS-III, FIM, and VAS for lumbar pain rating at hospital admission (T0), at hospital discharge (end of neurorehabilitation - T1), and 6 months later (T2). At T0, the evaluations were completed by an EMG study of trunk muscles activation.

Axial disorders are frequent complications of Parkinson's Disease (PD) and they can lead to postural deformities and balance impairments. The most prevalent postural disorders of PD are represented by camptocormia, antecollis, scoliosis, and Pisa syndrome (PS).

The pathogenesis of these disorders has not yet been completely elucidated, and it is characterized by a complex interlacement between central and peripheral mechanisms.

The first report of PS dates back in 1972, when Ekobm described a case series of three patients who developed a lateral trunk flexion in close temporal relationship with neuroleptics assumption. The roster of drugs associated with an acute/subacute onset of PS is huge and constantly growing, and it includes: antidepressants (mirtazapine, sertraline), cholinesterase inhibitor (rivastigmine, galantamine, donepezil), neuroleptics (tiapride, clotiapine, clozapine, aipiprazole, butyrophenone, paliperidone, quetiapine), dopamine agonists, (pramipexolo, ropinirole, pergolide), lithium, valproic acid, and betahistine. Nonetheless, PS was described as well in patients with neurodegenerative disorders, and in particular in PD and parkinsonism without drugs exposure.

The prevalence of PS is around 8.3% (9.3% in women and 6.4% in men) when calculated in a psychogeriatric population, and it is 8 to 8.8% when populations of PD patients were considered.

Formal diagnostic criteria for PS are not available, indeed the diagnosis is based on the clinical features:

• presence of a lateral flexion of the trunk with a homogenous angle between sacrum and spinous process of the 7th cervical vertebra;
• association of an ipsilateral rotation of the trunk around the sagittal axis that leads to a higher and anterior position of the shoulder contralateral to the side of trunk deviation;
• the postural disorder worsens during standing position, sitting position and gait;
• the postural disorder improves in supine position;
• disregard of patients for the postural disorder.

The fluctuation of the postural alteration during static (supine position vs. upright standing position) and dynamic conditions appears to be crucial to differentiate PS from scoliosis.

A lateral trunk deviation of at least 10° is commonly accepted for the diagnosis, although higher or lower cut-offs were used in the past. According to the degree of the lateral trunk flexion, PS can be further divided in mild (less than 20°) or severe (more than 20°) phenotypes.

Patients with PD and PS showed some typical clinical and demographic features when compared to PD patients without postural alterations: they are older, PD is longer in duration, more severe and with a more pronounced asymmetry of motor symptoms. The parkinsonian symptoms involving the upper limbs as well as gait impairment are more severe in PD patients with PS. Moreover, they are characterized by a higher incidence of falls, arthrosis, osteoporosis, orthopedics diseases and pain, specifically lumbar and lower back pain, which is reported in up to 75% of patients with PD and PS.

The management of PS is still a challenge for the physician. PS poorly responds to antiparkinsonian drugs, although there aren't studies specifically designed to assess this topic.

Neurorehabilitation represents one of the fundamental approach to PD and to postural disorders in general, not only for the treatment of the motor symptoms itself but also to improve quality of life and autonomy in the activity of daily living.

The study was a randomized, double-blind, controlled trial aimed to assess the efficacy of five daily sessions of bi-hemispheric t-DCS in add-on to an in-hospital neurorehabilitation protocol in patients affected by PD and PS.

At hospital admission (T0 - baseline), all patients underwent complete neurological, general and functional examinations by a Neurologist with expertise in movement disorders and neurorehabilitation. Patients who fulfilled inclusion and exclusion criteria underwent a baseline kinematic analysis of trunk movement. Patients with at least 10° of lateral trunk flexion completed the baseline evaluations with a dynamic electromyographic (EMG) study of trunk muscles, and with administration of a set clinical scales for the evaluations of motor disability, functional independence, and lumbar pain.

After that, patients were randomly assigned to "t-DCS" or "sham" treatment, and they started the double-blind phase of the study. The 5-day t-DCS/sham treatment was delivered in 5 daily consecutive sessions, starting from the first Monday after hospital admission. In parallel to neuromodulation, all patients were treated with a standardized 4-week rehabilitation programme. The kinematic analysis of trunk movement as well as the administration of the set of questionnaires were repeated at the end of the 4-week rehabilitation programme (T1 - hospital discharge), and 6 months after discharge (T2).

The randomization was performed according to a block randomization method. A unique randomization list was generated before enrollment with the following parameters: 6 blocks; 6 patients per block (3 for t-DCS group, and 3 for sham group).

All the patients were treated with an optimized and individualized anti-parkinsonian therapy, which dose and regimen were kept stable during the overall study period. All the evaluations were performed in the morning, and always in an ON phase.

The specific and standardized in-hospital rehabilitation programme was focused on the rehabilitation of the trunk postural disorder. All patients were treated with 90-minute daily sessions, 6 days a week (Monday to Saturday) for four weeks.

Each session was structured as follow:

• 10 minutes of cardiovascular warm-up activities: intersegmental coordination exercises; exercises to release shoulder and pelvic girdle; pelvic anteversion and retroversion movements to improve diaphragmatic respiration (supine position); breathing exercises to promote expansion of the chest;
• 15 minutes of stretching exercises: exercises to stretch the muscles of the posterior kinematic chain; exercises to stretch the pectoralis muscles; exercises to stretch the ischio-cruralis muscles; assumption of the prone position, sitting on heels and stretching the arms out in front; the "bridge" exercise to stretch the muscles of the anterior abdominal wall, glutei, quadriceps and hamstring; exercises to stretch lumbar muscles (in the supine position, each knee, in turn, is brought to the chest);
• 15 minutes of strengthening exercises in a functional context: exercises to strengthen the dorsal muscles (arms extended and hands outstretched as though to take something); lateral bending (arms lying along the body and hands reaching down as though to pick up something); stretching using the wall bars;
• 20 minutes of gait training: overground gait training (forwards, backwards, and lateral); walking on the spot;
• 15 minutes of balance training: path with obstacles; balance exercises performed in order of difficulty (heel-to-toe walking, lateral walking crossing the legs, walking along a path on surfaces of different texture);
• 15 minutes of relaxation exercises: intersegmental coordination exercises; segmental passive mobilization (until maximum joint range of motion is reached); breathing exercises to promote expansion of the chest.

Inclusion Criteria:

• age between 18 and 80 years;
• Hoehn and Yahr stage between II and III;
• Mini-Mental State Examination score above 24;
• lateral trunk flexion of at least 10° at baseline.

Exclusion Criteria:

• history of major psychiatric or other neurological conditions;
• history of back surgery, tumors or infections of the spine, intradural or extradural hematoma, ankylosing spondylitis, spinal stenosis;
• history of idiopathic scoliosis;
• botulin toxin treatment in the previous year;
• any change in dose or regimen of the anti-parkinsonian therapy in the last month before enrolment.

Thirty patients affected by Parkinson' Disease (PD) and Pisa Syndrome (PS) were consecutive enrolled among those attending the Neurorehabilitation Department of the IRCCS Mondino Foundation (Pavia, Italy). Idiopathic PD was diagnosed according to the Movement Disorders Society clinical diagnostic criteria for PD. Pisa syndrome was clinically diagnosed according to the following criteria:

• a lateral flexion of the trunk with a homogenous angle between sacrum and spinous process of the 7th cervical vertebra;
• an ipsilateral axial rotation of the trunk around the sagittal axis, that leads to a higher and anterior position of the shoulder contralateral to the side of trunk deviation;
• the worsening of the postural disorder during standing position, sitting position and gait;
• the improvement of the postural disorder in supine position.

Transcranial direct current stimulation (t-DCS) was delivered by an expert technician (V.G.) that was not otherwise involved in the management of the patients. The managing physician as well as the physiotherapist were instead blind to the type of stimulation.