AIM The striatal-to-occipital ratio (SOR) is commonly used as an analytical parameter in L-3,4-dihydroxy-6-18F-fluorophenylalanine (FDOPA) PET studies. It has been shown to be useful in differentiating idiopathic Parkinson's disease (IPD) patients from healthy individuals. We assessed the performance of SORs and subregional ratio of striatal-to-occipital ratios (RSORs) in the clinical assessment of nigrostriatal dopaminergic function for differentiating typical IPD from atypical parkinsonian disorders (APD). MATERIALS AND METHODS A total of 117 patients referred from movement disorder clinics in speciality neurology centres underwent an FDOPA PET study and were kept under follow-up for at least 2 years. Sixty-five patients (43 IPD and 22 APD) completed the 2-year follow-up and were included in the final analysis. Their PET images were spatially normalized to occipital counts and analysed with three striatal subregional regions of interest (caudate, anterior putamen and posterior putamen) and two occipital regions of interest. The RSORs of the caudate and posterior putamen, the caudate and anterior putamen, the caudate and whole putamen and the anterior putamen and posterior putamen were also calculated and compared between the IPD and APD groups using the t-test. RESULTS The P values for these SORs were found to be insignificant between IPD and APD patients (caudate: 0.1325; anterior putamem: 0.5469; and posterior putamen: 0.9835). However, the RSORs of the caudate and posterior putamen showed significant differences between these two populations of patients. CONCLUSION The SOR method is already known to be a good diagnostic tool to differentiate between IPD patients and the normal population. SOR, however, fails to distinguish IPD from APD patients, and hence the RSOR of the caudate and posterior putamen can be utilized to differentiate between them.