Table 1 A detail summary of risk factors for FOG development in Parkinson’s disease

CSF Aβ42 was a predictor of FOG in patients with early PD. PIGD score, caudate DAT uptake, and CSF Aβ42 together could predict FOG within 4 years after diagnosis of PD (AUC = 0.755).

PPMI database (393 early do novo PD without FOG at baseline)

CSF (Aβ42, α-synuclein, t-tau, p-tau, Aβ42/ t-tau); motor; non-motor (olfactory, sleep, cognition, depression, anxiety, autonomic function); DAT imaging (striatal region)

Median 4.0 years

MDS-UPDRS item 2.13 or item 3.11 ≥ 1

[16]

DAT uptakes in the caudate nucleus and putamen predicted the development of FOG. Male sex, higher PIGD score, and lower MoCA score were also significant predictors of FOG.

PPMI data (390 early do novo PD without FOG at baseline)

DAT imaging (striatal region); motor; cognition

Median 4.0 years

MDS-UPDRS item 2.13 or item 3.11 ≥ 1

[17]

FOGQ total score and the anxiety score were the strongest predictors, and using only these two factors could significantly predict FOG in the next 15 months with 82% accuracy.

221 PD in which 88 patients had FOG at baseline

Motor; non-motor (cognition, anxiety, depression, sleep); medication use

Mean 14 months

FOGQ item 3 ≥ 1

[18]

Depressive, gait speed and UPDRS-III (off vs. on) were the independent predictors of future FOG.

57 PD patients without FOG at baseline

Motor; non-motor (sleep, cognition, autonomic, depression, and others); medication use

Mean 5 years

NFOGQ (item1 = 1) and objective observation

[19]

Increased risk: Onset of PD with a gait disorder; higher scores of rigidity, postural instability, bradykinesia and speech; and longer disease duration; absence of tremor. Decreased risk: Deprenyl treatment

DATATOP data (800 early PD in which 57 patients had FOG at study entry)

Motor; non-motor (speech, cognition, depression); initial symptoms

Mean 14 ± 5 months

UPDRS II-item14 ≥ 1

[20]

Motor fluctuations, higher levodopa dose were independent risk factors; none of the cardinal features independently predicted FOG.

232 PD without FOG at baseline

Motor and motor complication; non-motor (psychosis (UPDRS I item2, hallucinations or delusions), cognition); medication use

12 years

UPDRS II-item14 ≥ 1

[21]

Lower education, akinetic-rigid style, not using dopamine receptor agonists, sleep disorders (insomnia); cognitive disturbances were predictors of FOG.

248 early PD without FOG at baseline

Motor; non-motor (anxiety; depression); medication use;

3 years

FOGQ item 3 ≥ 1 and objective observation

[22]

Longer disease duration, visuospatial function deterioration, onset in lower limbs, presence of festination, falls, and hallucinations were independent predictors of FOG.

225 PD without FOG at baseline

Motor (including festination and fall); non-motor (anxiety, depression, cognition); medication use;

3 years

FOGQ item 3 ≥ 1 and objective observation

[23]

Baseline processing speed, learning and daytime sleepiness were predictive of FOG.

PPMI database (50 PD + FOG, and 50 PD-FOG at the fourth year)

Motor; non-motor (cognition, anxiety, depression, sleep)

4 years

MDS-UPDRS item 2.13 ≥ 1

[24]

A more severe depletion of presynaptic dopamine (low 123I-FP-CIT binding in the putamen and striatum) in early PD predicted FOG

41 early PD without FOG at baseline

Motor (including falls) and motor complication; DAT imaging (striatal region); medication use

9.51 ± 3.18 years

UPDRS II-item14 ≥ 1

[25]

PD group with moderate to severe WMH showed a higher risk of developing FOG (HR, 3.29; 95% CI, 1.79–6.05; P < 0.001) than the PD patients with minimal WMH.

268 patients with de novo PD without FOG

MRI WMH; motor (UPDRS-III, phenotype); non-motor (olfactory, cognition, depression) DAT imaging; medication; vascular risk factors

>  3 years

Inquiry and observation

[26]