Evidence for exercise on FOG in PD Moran Gilat PhD Department of - PowerPoint PPT Presentation

Evidence for exercise on FOG in PD Moran Gilat PhD Department of Rehabilitation Sciences Supported by European Commission Marie- Skłodowska Curie Actions Fellowship

Disclosures • None 2

Freezing of gait (FOG) Common -- Episodic -- Worsening -- Disabling Triggers Clinical Motor Non-motor • Turning • Gait variability • Executive dysfunction • Disease severity • Narrow passages • Worse in OFF • Smaller steps • Sensory deficits • Gait initiation • Postural instability • Anxiety • Extra-nigral pathology • Dual-tasking Schaafsma et al. 2009; Nutt et al. 2011; Ehgoetz Martens et al. 2014; Bohnen et al. 2014; Bekkers et al. 2018 3

Evolution of freezing Gilat et al. 2020 in preparation; Bohnen et al. 2014; Peterson and Horak 2016; Gilat et al. 2019 4

Why bother with PT? Medications / DBS Does PT work? Physical Therapy No FOG Mild FOG Severe FOG 5

Research Questions 1. How much evidence is there on the effect of physical therapy for reducing FOG? -Immediate effect -Long term effect 2. What type of intervention should we offer PD with FOG? 6

Physical therapy for FOG in PD 1 FOG-specific FOG-correlates Generic exercise Aims Aims Aims -  Physical & Mental fitness 2 - Reduce FOG episodes - Train (non-)motor correlates -  Sleep quality 3 - Prepare for upcoming FOG → More resilience against FOG -  Dopamine release striatum 4 -  Motor symtom progression 2,5 Examples Examples Examples • • • Cueing Balance/Posture/APA Dancing • • • Gait feedback Cognitive training Treadmill • • • Action observation Motor-Cognitive training Aquatic exercise • • Fall prevention Conventional PT • Etc. 1=Gilat et al. 2020 In prep; 2=Schootemeijer et al. 2020; 3=Cristini et al. 2020; 4=Sacheli et al. 2019; 5=van der Kolk et al. 2019 7

FOG-specific example: Cueing Hypokinetic Gait and FOG in PD “ “ χ Visual Cueing Auditory Cueing Visual Cueing -Stripes -Laser shoe - On demand χ +20% χ +20% Nieuwboer et al. 2007; Fietzek et al. 2014; Spildooren et al. 2017; Ferraye et al. 2016; Ginis et al. 2017 8

FOG-specific example: Action Observation Pelosin et al. 2010; 2018; Agosta et al. 2017; Mezzarobba et al. 2017 9

FOG-correlates examples Motor = Balance / Weight-shifting / Turning / Gait training Cognitive training Motor-Cognitive training RED GREEN RED GREEN BLUE RED RED GREEN BLUE RED GREEN BLUE BLUE RED GREEN BLUE GREEN BLUE . . Walton et al. 2018 King et al. 2020; Clerici et al. 2019; Silva-Baptista et al. 2020 Bekkers et al. 2020 10

Systematic Search Databases -PubMed -EMBASE -MEDLINE -Web of Science -Google Scholar Search conducted 3 rd August 2020 • • Screening and data extraction by two researchers INCLUSION CRITERIA 1) RCT of training/exercise with FOG as an outcome 2) Training at least two days 3) N>10 4) Any language 5) Any date PROSPERO Registration: 42019123882 Gilat et al. 2020 in preparation 11

Meta-analysis FOG outcomes used • RevMan (v5.3) (N)FOGQ 43 MDS-UPDRS II 1 Analysis of final measures MDS-UPDRS III 1 Contrasts: 1 FOG diary 1) Effect of any type of training/exercise -Passive control 4 -Active control FOG-score 2 1) Subgroup (Categories) FOG-ratio 1) Long-term (follow-up) effect %FOG 1 2) Effect in freezers only Gilat et al. 2020 in preparation 12

1) Any training/exercise *� *� 42 studies *� Total 1838 PD patients -933 Intervention -905 Control ES=-0.37 [-0.51, -0.22] , p<0.00001 , I 2 =52% - 3 outliers ES=-0.24 [-0.35, -0.14] , p<0.00001 , I 2 =9% Gilat et al. 2020 in preparation 13

1) Type of Control Active control Passive control *� *� ES=-0.30 [-0.49, -0.11] , p=0.002 , I 2 =30% ES=-0.20 [-0.32, -0.08] , p<0.001 , I 2 =0% Gilat et al. 2020 in preparation; Cosentino et al. 2019 14

2) Subgroups: FOG-specific Aimed at the alleviation of FOG episodes 12 studies *� Total 807 PD patients -403 Intervention -404 Control ES=-0.35 [-0.56, -0.13] , p<0.01 , I 2 =45% - 1 outlier ES=-0.24 [-0.40, -0.07] , p<0.01 , I 2 =14% Gilat et al. 2020 in preparation 15

2) Subgroup: FOG-correlates Aimed at the underlying correlates of FOG 13 studies *� Total 422 PD patients -263 Intervention -159 Control ES=-0.59 [-0.95, -0.24] , p<0.01 , I 2 =72% - 1 outlier ES=-0.40 [-0.64, -0.16] , p<0.01 , I 2 =36% Gilat et al. 2020 in preparation 16

2) Subgroup: Generic Exercise Generic exercises *� 16 studies Total 1838 PD patients -267 Intervention -242 Control ES=-0.20 [-0.39, -0.01] , p=0.04 , I 2 =11% - 1 outlier ES=-0.14 [-0.32, 0.04] , p=0.80 , I 2 =0% Gilat et al. 2020 in preparation 17

2) Subgroups: Effect sizes FOG specific FOG correlates Generic exercise ES=-0.24, p<0.01 ES=-0.40, p<0.01 ES=-0.14, p=0.80 Effective Not effective Gilat et al. 2020 in preparation 18

Model for PT No� FOG� Occasional� FOG� � Frequent� FOG� Present � C� Generic Excercise Future � B� FOG Correlates � A� FOG Specific Disease duration Gilat et al. 2020 in preparation 19

3) long-term effect Retention Category period 15 studies 8w 2 8w 2 4w 1 4w 2 Total 520 PD patients 4w 1 -257 Intervention 4w 2 4w 3 -263 Control 4w 3 12w 3 *� 24w 2 Follow-up 1-6 months 12w 1 ES=-0.16 [-0.36, 0.03] , p=0.10 , I 2 =16% 4w 1 4w 1 24w 2 - 1 outliers 24w 1 ES=-0.08 [-0.27, 0.10] , p=0.36 , I 2 =0% Without exercise: ES=-0.09, p=0.11 Gilat et al. 2020 in preparation 20

3) long-term effect Our meta-analysis: • All trials (n=14) → p=0.36 • FOG related (n=12) → p=0.11 Cosentino et al. (2019) 1 • Exercise (n=2) → p=0.93 Retention possible, but impaired in freezers 2 • PT (n=8) → p<0.001 -Action observation (n=4) → p=0.02 → Target these patients EARLY 2 -Cueing (n=2) → p=0.78 -Treadmill / Aquatic → In future, target those at risk for FOG 3 1=Cosentino et al. 2019; 2=Nieuwboer et al. 2004; 3=D’Cruz et al. 2020 21

Early start and long-term engagement Not self-evident in this population : Predictors of poor adherence 1 - FOG - Falls - Mental health (anxiety & depression) Multidisciplinary care - Cognitive decline - Pain Lack of exercise partner 2 - Predictors of good adherence 1 Start EARLY - Prior experience with exercise 1 = Allen et al. 2015; 2 = Zaman et al. 2019 22

Early start and long-term engagement Good news : Increasing variety of FOG-based interventions on offer → Currently low availability, but rising N studies 8 → Supervision needed 7 6 → Consider Group classes 1 5 4 3 2 1 0 ‘17 -18 ‘19 -20 ‘11 -12 ‘13 -14 ‘15 -16 publication year 1=King et al. 2015 23

Limitation 1 Not all RCT’s were directly aimed at FOG, with many also including non -freezers Category Freezers only 2 1 14 studies 1 *� 2 1 2 Total 1838 PD patients -314 Intervention -296 Control 1 1 1 2 2 1 ES=-0.53 [-0.87, -0.18] , p=0.002, I 2 =73% 2 1 - 1 outlier ES=-0.32 [-0.55, -0.09] , p=0.007 , I 2 =38% Gilat et al. 2020 in preparation 24

Limitation 2 A:� Layout� of� the� ‘ FOG-scoring ’ � template� Use of the N(FOGQ) as an outcome + Easy and quick to use - Subjective scale → recollection bias - Minimal detectable change ~10 points! 1 Fully automated IMU at-home assessment %FOG video B:� Example� of� the� output� FOG score FOG ratio FOG diary (N)FOGQ 2 Download for FREE at → morangilat.com UPDRS single item 1=Hulzinga et al. 2020; 2=Gilat 2020 25

TAKE HOME MESSAGE ➢ Exercise is very important, but FOG reduction not to be expected ➢ At FOG onset, offer PT targeting motor- and non-motor correlates to increase resilience ➢ When FOG becomes frequent, offer FOG-specific PT interventions to reduce its impact  Start EARLY  Ensure supervision and consider group-classes  PD with FOG will need extra motivation and follow-up ✓ Targeted PT has a significant effect on the impact of FOG in PD! 26

Thank you Questions? moran.gilat@kuleuven.be