Higher serum cortisol levels appear to be associated with stress that can overlap or manifest anxiety, fatigue, depression, and sleep dysfunction. These are common and intrusive non-motor symptoms of Parkinson's disease (PD). Thus, stress has been proposed to mediate Parkinson's disease development, and cortisol has been suggested as a biomarker for the generation of stress-related symptoms in Parkinson's disease. This study describes sensitive and robust disposable pipette extraction (DPX) and ultra-efficient liquid chromatography tandem mass spectrometry (UHPLC-MS/MS) method to determine cortisol and cortisone (as potential endocrine biomarkers for Parkinson's disease) in 24-h urine and saliva samples obtained from Parkinson's disease patients. Important parameters on DPX extraction were optimized to achieve the best extraction recovery and cleanup efficiency. The proposed method was linear from 0.5 (lower limit of quantification) to 500ngmL-1 for cortisol and from 3.0 (lower limit of quantification) to 500ngmL-1 for cortisone. To determine whether urinary cortisol and urinary cortisone are adequate as biomarkers to evaluate the level of anxiety in patients suffering from Parkinson's disease, twenty-nine Parkinson's disease patients (18 with anxiety and 11 without anxiety) were selected for urine analysis. Based on the obtained results, 24-h urine samples obtained from Parkinson's disease patients with anxiety had higher cortisone levels than samples obtained from healthy controls. Receiving operating curves (ROC) analysis, which presented the area under the ROC curve (AUC = 0.733), showed that urinary cortisone levels (µg/24-h urine) were sensitive (56.3%) and specific (93.3%) for distinguishing Parkinson's disease patients with anxiety from healthy controls. In terms of salivary results, PD patients' samples taken 30min after waking up had higher cortisol and cortisone levels than healthy controls, while their samples taken at night had lower cortisol and cortisone levels.