To date, there are no definitive biomarkers for Parkinson's disease (PD) diagnosis. The detection of cerebrospinal fluid (CSF) alpha (α)-synuclein in PD patients has yielded promising but inconclusive results. To determine the performance of CSF α-synuclein as a diagnostic biomarker of PD and whether CSF α-synuclein can discriminate PD from other neurodegenerative diseases, a systematic search of all relevant studies investigating reproducible CSF α-synuclein quantification methods was conducted in electronic databases. A total of 17 studies that included 3311 patients were included in this systemic review and meta-analysis. The mean CSF α-synuclein concentration was significantly lower in PD patients compared to normal/neurological controls [weighted mean difference (WMD) −0.31; 95% CI, −0.45, −0.16; p < 0.0001] and patients with Alzheimer's disease (AD) [WMD −0.15; 95% CI, −0.26, −0.04; p < 0.0001]. There was no significant difference between PD patients and dementia with Lewy bodies (DLB) patients [WMD −0.03; 95% CI, −0.16, 0.09; p = 0.58] or patients with multiple system atrophy (MSA) [WMD 0.05; 95% CI, −0.04, 0.13; p = 0.25]. Sensitivity and specificity of CSF α-synuclein in the diagnosis of PD was 0.88 (95% CI, 0.84–0.91) and 0.40 (95% CI, 0.35–0.45), respectively. The positive and negative likelihood ratios of CSF α-synuclein in the diagnosis of PD were 1.41 (95% CI, 1.24–1.60), and 0.29 (95% CI, 0.15–0.56), respectively. The corresponding summary receiver operating characteristic (SROC) curve showed an area under the curve (AUC) of 0.73. The concentration of CSF α-synuclein may be a biomarker for the diagnosis of PD. The use of α-synuclein alone however is not sufficient as a single biomarker and it must therefore be used in conjunction with other documented and reliable biomarkers.