Clinical value of machine learning-based interpretation of I-123 FP-CIT scans to detect Parkinson's disease: a two-center study.

PURPOSE: Our aim was to develop and validate a machine learning (ML)-based approach for interpretation of I-123 FP-CIT SPECT scans to discriminate Parkinson's disease (PD) from non-PD and to determine its generalizability and clinical value in two centers. METHODS: We retrospectively included 210 consecutive patients who underwent I-123 FP-CIT SPECT imaging and had a clinically confirmed diagnosis. Linear support vector machine (SVM) was used to build a classification model to discriminate PD from non-PD based on I-123-FP-CIT striatal uptake ratios, age and gender of 90 patients. The model was validated on unseen data from the same center where the model was developed (n = 40) and consecutively on data from a different center (n = 80). Prediction performance was assessed and compared to the scan interpretation by expert physicians. RESULTS: Testing the derived SVM model on the unseen dataset (n = 40) from the same center resulted in an accuracy of 95.0%, sensitivity of 96.0% and specificity of 93.3%. This was identical to the classification accuracy of nuclear medicine physicians. The model was generalizable towards the other center as prediction performance did not differ thereby obtaining an accuracy of 82.5%, sensitivity of 88.5% and specificity of 71.4% (p = NS). This was comparable to that of nuclear medicine physicians (p = NS). CONCLUSION: ML-based interpretation of I-123-FP-CIT scans results in accurate discrimination of PD from non-PD similar to visual assessment in both centers. The derived SVM model is therefore generalizable towards centers using comparable acquisition and image processing methods and implementation as diagnostic aid in clinical practice is encouraged.

Unilateral pallidotomy versus bilateral subthalamic nucleus stimulation in Parkinson's disease: one year follow-up of a randomised observer-blind multi centre trial.

BACKGROUND: To investigate whether STN stimulation is more efficacious than unilateral pallidotomy in advanced Parkinson's disease (PD) one year after surgery. METHOD: Thirty-four patients with advanced PD were randomly assigned to unilateral pallidotomy or bilateral STN stimulation. Outcome measures were parkinsonian symptoms in off and on phases (UPDRS 3), dyskinesias, functional status, Parkinson's disease quality of life questionnaire, the effects on separate symptoms, timed tests, patient diaries, dopaminergic drugs changes, adverse effects, and global outcome scale. Patients were assessed before surgery, six months and one year after surgery. The primary outcome measure was the off phase UPDRS 3 at six months follow-up. FINDINGS: The off phase UPDRS 3 score improved from 46.5 to 32 points in the pallidotomy patients and from 51.5 to 24 in the STN stimulation patients (p = 0.002). On phase UPDRS 3 and off phase Schwab and England functional scale improved significantly in favour of the STN stimulation patients. Dopaminergic drugs reduction was larger in the STN group although the difference between the treatment groups was not significant. One patient in each group had a major adverse effect. CONCLUSIONS: Bilateral STN stimulation is more efficacious than unilateral pallidotomy in advanced PD up to one year after surgery.

Chronic stimulation of the subthalamic nucleus increases daily on-time without dyskinesia in advanced Parkinson's disease.

We assessed the efficacy of chronic stimulation of the subthalamic nucleus (STN-DBS) in 20 patients with Parkinson's disease (PD) by means of clinical assessments and patient diaries 12 months after surgery. STN-DBS reduced the UPDRS part III off-medication score by 33%, and successively improved complete daily on-time without dyskinesia at 12 months significantly. In conclusion, our study demonstrates the efficacy of chronic STN-DBS on motor features in a selected population of advanced PD patients. In addition to clinical assessments, patients' diaries serve as an essential tool to evaluate the functional motor status after STN-DBS.

Disease progression continues in patients with advanced Parkinson's disease and effective subthalamic nucleus stimulation.

OBJECTIVES: Glutamate mediated excitotoxicity of the hyperactive subthalamic nucleus (STN) has been reported to contribute to nigral degeneration in Parkinson's disease (PD). Deep brain stimulation of the STN (STN DBS), in its role as a highly effective treatment of severe PD motor complications, has been thought to inhibit STN hyperactivity and therefore decrease progression of PD. METHODS: In a prospective two centre study, disease progression was determined by means of serial (18)F-fluorodopa (F-dopa) positron emission tomography (PET) in 30 patients with successful STN DBS over the first 16 (SD 6) months after surgery. RESULTS: Depending on the method of PET data analysis used in the two centres, annual progression rates relative to baseline were 9.5-12.4% in the caudate and 10.7-12.9% in the putamen. CONCLUSIONS: This functional imaging study is the first to demonstrate a continuous decline of dopaminergic function in patients with advanced PD under clinically effective bilateral STN stimulation. The rates of progression in patients with STN DBS were within the range of previously reported data from longitudinal imaging studies in PD. Therefore this study could not confirm the neuroprotective properties of DBS in the STN target.

The nigrostriatal dopaminergic system in familial early onset parkinsonism with parkin mutations.

Nigrostriatal dopaminergic function and cerebral energy metabolism were measured with PET in two brothers with early-onset parkinsonism caused by mutation of the parkin gene. Energy metabolism did not differ, but the nigrostriatal dopaminergic pattern was clearly different than that of sporadic PD. Thus parkinsonism in these two patients was shown to be pathophysiologically different than PD.

Familial parkinsonism with synuclein pathology: clinical and PET studies of A30P mutation carriers.

BACKGROUND: The authors identified the second known mutation in the alpha-synuclein(SNCA) gene, an alanine-to-proline exchange in amino acid position 30 (A30P), that cosegregates with the disease in one German family with autosomal dominantly inherited parkinsonism (ADP). The authors studied carriers of the A30P mutation to compare the phenotype of this mutation with idiopathic PD (IPD) and to assess nigrostriatal dopaminergic function in symptomatic and preclinical mutation carriers. METHODS: The pedigree of the A30P family spans five generations with five affected individuals. The authors performed detailed neurologic examinations followed by mutation analysis in 11 living individuals. In three mutation carriers, two individuals with definite PD and one person at risk for PD, they used L-[18]F-fluoro-3,4-dihydroxyphenylalanine (F-DOPA), [11]C-raclopride (RAC), and [18]F-fluorodeoxyglucose (FDG) PET to investigate presynaptic dopaminergic function, dopamine D2 receptors, and cerebral energy metabolism. The authors studied the cognitive functions of carriers of the A30P mutation using neuropsychological screening. RESULTS: PET studies revealed striatal presynaptic dopaminergic alterations consistent with sporadic IPD in two affected family members and no evidence for nigrostriatal dopaminergic dysfunction in one presymptomatic mutation carrier. Neuropsychological testing in four mutation carriers provided evidence for cognitive impairment as a frequent and early symptom of the A30P mutation; this is also supported by regional cerebral energy metabolism alterations in the clinically presymptomatic subject. CONCLUSIONS: The phenotype of the A30P mutation in the SNCA gene is similar to that of sporadic IPD, including a high variability of the age at disease onset, ranging from 54 to 76 years. The follow-up of presymptomatic carriers of the A30P mutation may give insight into preclinical disease stages and early manifestations of PD.

Valproic acid toxicokinetics: serial hemodialysis and hemoperfusion.

The toxicity and pharmacokinetic properties of a drug determine whether hemodialysis and/or hemoperfusion are indicated in acute intoxications. Valproic acid is considered unremovable by hemodialysis because of the high protein binding of 90%-95%. A 27-year-old male with a history of seizures was admitted to the emergency room because of coma, hypernatriemia, and respiratory failure caused by an intoxication with a large dose of valproic acid. At admission, the plasma valproic acid level was 1414 mg/L (9.9 mmol/L) (therapeutic range: 50-100 mg/L (350-700 micromol/ L). The anion gap was 26 mmol/L (normal <12-14 mmol/L) and corresponded fairly well with this valproic acid level. Because of the potential toxicity of this high valproic acid level serial hemodialysis and hemoperfusion was performed. The first session was done with a charcoal column and the second session with a resin column. The patient recovered during the course of treatment. The valproic acid plasma clearances during treatment were: 80 mL/min (hemodialysis); 40 mL/min (hemoperfusion by charcoal) and 80 mL/min (hemoperfusion by resin, only in the first hour). The protein binding of valproic acid in plasma was only 32% at the start and was 54% at the end of the two sessions. In this specific case of a severe valproic acid intoxication, saturated protein binding resulted in an increased fraction of unbound valproic acid. This made hemodialysis an effective treatment, while hemoperfusion was relatively less effective because of saturation of the column. In conclusion, the toxicokinetics of valproate are quite different from the pharmacokinetics at therapeutic levels. The anion gap and protein binding are important parameters in toxicokinetics.