Dissociative identity disorder (DID) is characterized by the existence of two or more separate and distinct personalities on an ongoing basis. Although the incidence of DID has been increasing in Japan, dental treatments for such patients remain rare. Here, we report our experience providing general anesthesia to a patient with DID and generalized anxiety disorder.
A 30s woman had been diagnosed as having DID at the age of 24 years, and her condition was being managed with medication. Although more than nine personalities had appeared as of the presently reported treatment, her original personality had not appeared for several years. She had numerous dental caries and an extreme fear of dental treatments. Therefore, we planned to perform dental treatments under general anesthesia.
Anesthesia was maintained with 3.0 μg/ml of propofol and 0.05-0.1 μg/kg/min of remifentanil hydrochloride. Since DID patients can respond to anesthetics differently than healthy adults, we used a BIS monitor to ascertain the depth of the anesthesia. Furthermore, the dentists, nurses, and other staff members shared information during the perioperative period. As a result, while changes in the patient’s personality were observed, no serious mental disturbances occurred.
Objectives: To access medical specialists by health specialists working in the regional health field, the possibility of utilizing the voice approach for dissociative identity disorder (DID) patients as a health assessment for medical access (HAMA) was investigated. The first step is to investigate whether the plural personae in a single DID patient can be discriminated by voice analysis.
Methods: Voices of DID patients including these with different personae were extracted from YouTube and were analysed using the software PRAAT with basic frequency, oral factors, chin factors and tongue factors. In addition, RAKUGO story teller voices made artificially and dramatically were analysed in the same manner. Quantitive and qualitative analysis method were carried out and nested logistic regression and a nested generalized linear model was developed.
Results: The voice from different personae in one DID patient could be visually and easily distinquished using basic frequency curve, cluster analysis and factor analysis. In the canonical analysis, only Roy’s maximum root was <0.01. In the nested generalized linear model, the model using a standard deviation (SD) indicator fit best and some other possibilities are shown here.
Conclusions: In DID patients, the short transition time among plural personae could guide to the risky situation such as suicide. So if the voice approach can show the time threshold of changes between the different personae, it would be useful as an Access Assessment in the form of a simple HAMA.