Validação dos critérios diagnósticos de psicose associada à doença de Parkinson propostos pelo Ninds-Nimh

• Ano de defesa: 2017
• Autor(a) principal: Gomes, Roberta Borges [UNIFESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Paulo (UNIFESP)
• Palavras-chave em Português: Parkinson'S disease; Psichosis; Halucinations; Diagnostic criteria; Doença de Parkinson; Psicose; Alucinações; Critérios diagnósticos; Psicose da doença de Parkinson
• Link de acesso: https://sucupira.capes.gov.br/sucupira/public/consultas/coleta/trabalhoConclusao/viewTrabalhoConclusao.jsf?popup=true&id_trabalho=5454608; http://repositorio.unifesp.br/handle/11600/50032

Parkinson's disease (PD) is a neurodegenerative disorder characterized predominantly by motor manifestations. However, psychotic symptoms affect more than 50% of patients in the more advanced stages, giving a worse prognosis. Most studies of psychiatric complications in PD have used generic assessment tools. The absence of well-defined criteria for psychosis associated with PD (PAPD) compromises epidemiological and pathophysiological studies and constitutes an obstacle to the establishment of appropriate treatment. In 2007, the NINDS-NIMH proposed five criteria for PAPD (PAPD-NINDS), among which, the patient should have at least one of the following symptoms for diagnosis: hallucination, delusion, illusion and false sense of presence. These criteria, however, lacked clinical validation. Objective: To validate the diagnostic criteria for Psychosis Associated with PD proposed by NINDS (PADP-NINDS) in a cohort of Brazilian patients with PD. Method: A cross-sectional study with consecutive evaluation of patients with PD from the Movement Disorders Clinics of Santa Marcelina Hospital at São Paulo/BR, between September 2012 and December 2014. All patients underwent evaluation by a neurologist expert in movement disorders which applied the criteria of PDAP-NINDS and other scales (UPDRS, Hoehn and Yahr, Schwab and England, PDQ-39, sleep evaluation scales, non-motor symptoms scale in PD, Pfeffer Scale, ACE- R, MEEM, MOCA, SCOPA-PC). Patients also underwent formal neuropsychological evaluation for the investigation of dementia in PD according to the criteria proposed by MDS. The gold standard assessment and diagnosis for PADP was performed by a psychiatrist through a structured clinical interview, which is recommended by DSM-IV. Both the neurologist and the psychiatrist, as neuropsychologist were blinded to the evaluations of other colleagues. Patients were also blinded to the results of their evaluations. Sample characterization was carried out through the mean summary measures, and the inferential analysis was used Student's t-test for related samples in the evaluation of the PDAP-NINDS Criteria in relation to numerical scales. The Fisher's exact test was used to compare the criteria in relation to scales with variables. Patients were further divided into 3 subgroups, according to agreement between the Criteria and DSM-IV: psychosis (diagnosis of psychosis by DSM-IV), subsyndromic (symptoms of psychosis but did not had diagnosis of psychosis by DSM-IV), Non-psychotic (without symptoms of psychosis). Each subgroup was treated as a dependent variable. A series of univariate analyzes were performed to compare independent variables and dependent variables and a MANOVA was performed to investigate the difference of the dependent variables between each subgroup. Cohen's kappa calculation for diagnosis using the NINDS criteria as well as for the presence of each individual characteristic symptom was performed. The proposition of a model in which the symptoms of the NINDS criteria would constitute a continuous scale, in which each criterion has a weight was created, after which the ROC and AUC curve was calculated. Results: Patients were male in majority (66.7%), mean age 63.9 years (SD = 10.7), age 55.9 years (SD = 10.7) at onset of disease, duration of disease of 7.9 years (SD = 4.8), H & Y of 2.8, UPDRS of 68.95 and dopaminergic equivalent of 557.4. Only 14.7% of the sample met criteria for dementia. Regarding the presence of psychotic symptoms, 51% of the sample had no symptoms and the other 49% had diverse combinations of psychotic symptoms. According to a psychiatric evaluation based on DSM-IV, 16 patients were had psychosis (15.2%). In the subgroup analysis, individuals with psychosis were mostly men (93.8%), had higher H & Y (4 and 5) and more dementia in comparison to subjects of the other subgroups (68.8% in the psychosis group and 4.7% % in the group of non-psychotic patients). Multivariate analysis showed statistical significance for presence of dementia, Epworth daytime sleepiness score, and S&E daily life activity score. When the diagnosis of psychosis was evaluated by the DSM-IV and the PAPD-NINDS criteria in the originally proposed way had kappa = 0.30. When suggesting a model that assigns two points to the DELUSION symptom and one point for each of the other PAPD-NINDS symptoms, choosing the cut-off point equal to or greater than 3, the reliability of the criteria improved significantly, with the accuracy above 90%. Considering this model for PAPD diagnosis, in multivariate regression analysis, only the Epworth score and the diagnosis of dementia showed statistical significance, the latter being the most relevant predictive factor in logistic regression analysis (OR 11.97; p 0.027). Conclusion: In the present study, we found that the NINDS-NIMH criteria proposed for the diagnosis of PAPD have poor agreement with the gold standard method in the diagnosis of mental disorders. They were shown to be excessively inclusive, with high sensitivity to the detriment of specificity. We suggest a modified NINDS model that showed 94% sensitivity and 91% specificity. Males, excessive daytime sleepiness and the diagnosis of dementia are positive predictive factors for PAPD.