Even if several studies have been carried out to evaluate temperament and personality traits among PD patients, PeDs, defined according to the DSM-IV, have been scarcely investigated. PeDs represent a quite common condition in the general population, and in particular prevalence of OCPeD in the general population has been estimated to be around of about 8.0% in a very large population survey recently carried out 
. In agreement with this data we found a close frequency of OCPeD in our control group (10%) 
Although most patients with a PeD do not require specific treatment, some more severe cases may improve with pharmacological and/or psychotherapeutic treatment 
. On these grounds, our survey firstly has investigated the prevalence of PeDs defined according to the DSM-IV in a PD population. Overall we found an higher frequency of PeDs in PD patients respect to a control group, mainly due to a very high frequency of OCPeD among PD patients (40%) and, in part, also to a higher frequency of depressive PeD (14% in PD patients and 4% in controls subjects).
According to the DSM-IV, obsessive compulsive personality disorder (OCPeD) is defined as a pervasive pattern of preoccupation with orderliness, perfectionism, and mental and interpersonal control at the expense of flexibility, openness, and efficiency 
. It should be underlined that the characteristics of OCPeD, as defined by the DSM-IV criteria, appear to overlap with the “parkinsonian personality” consistently reported in literature over the time 
. Our data suggest that such characteristics could be considered not just a trait, as it has been commonly described, but as a well defined disorder that could be diagnosed according to worldwide accepted diagnostic criteria.
Literature data have also suggested a possible increased frequency of Obsessive-Compulsive Disorders (OCD, noted on Axes I according to the DSM-IV) among PD patients, but to date, probably due to methodological differences across the studies, this association remains controversial 
. On the other hand also the relationship between OCD and OCPeD has been a subject of interest and controversy. OCPeD, in fact, has been regarded as a precursor to OCD, but to date, even if there are some overlap between these two conditions, how closely they are remains uncertain 
. PD is characterized by dysfunction in the fronto-basal ganglia circuitry and a similar circuitry has also been implicated in the pathophysiology of OCD. It is believed that the higher incidence of obsessive-compulsive symptoms in PD is due to the involvement of a shared circuitry 
In agreement with some literature evidences in our sample, frequency of OCPeD was significantly higher among PD patients aged 60 years or above. Prevalence studies addressing specific personality disorders, in fact, have shown that within different subpopulations personality pathology of clusters C, such as OCPeD, are quite common in older adults 
. It should be underlined that, although personality disorders have historically been considered as stable over time, clinical presentation of these disorders may change across the life-time including in the elderly. Personality disorders, in fact, may attenuate, re-emerge or appear de novo according to the social context. 
Concerning the other PeDs, the second most common disorder in our PD population was the Depressive PeD. Depressive personality disorder is a controversial psychiatric diagnosis that denotes a personality disorder with depressive features. Depressive personality disorder was added to DSM-IV’s appendix B (Criteria Sets and Axes Provided for Further Study) amid controversy and the main issue was whether it could be differentiated conceptually and empirically from dysthymia and major depression 
. Considering the well known high prevalence of depression among the PD patients, we are aware that in some case distinguishing between these two conditions can be difficult.
Finally also avoidant PeD (characterized by a pervasive feelings of social inhibition and social inadequacy, extreme sensitivity to negative evaluation and avoidance of social interaction) and dependent PeD (characterized by a pervasive psychological dependence on other people) were more common among PD patients than controls. According to the DSM-IV 
both avoidant and dependent personality disorders, along with the obsessive-compulsive one, are grouped in the cluster C. Even if the few number of events in these categories do not allow us to exclude the possible role of chance, the possible association between PD and both avoidance and dependent PeD could be explained by an high level of harm avoidance, that according to Cloninger’s model 
, is the temperament trait characterizing PD patients 
. A recent study, in fact, has highlighted the relationships between dimensional and categorical approaches of personality showing a positive correlations between harm avoidance and cluster C personality disorders 
We are aware that common limits of cross-sectional studies should be taken into account in interpreting our analyses. Selection of cases and controls generally represents the most important pitfall in case-control studies. As in several of case-control studies, also in our study PD patients were enrolled from a hospital setting. Cases selected from a hospital setting could include more severely affected patients not representing the total parkinsonian population and possibly resulting in a reduced generalization. However in our sample the Hoehn-Yahr stage (1.9±0.6) revels a mild or moderate stage of disease. Moreover we cannot be sure that the patient’s personality traits were evident before the onset of PD, in young adulthood and endured throughout life. To this reason, as recommended by the SCID-II guideline, subjects were specifically instructed to answer considering their entire life since young adulthood. Furthermore, during the interview, for each positive item of the SCID-II questionnaire, the examinator specifically asked if that behaviour was present since the adolescence 
. However the possible bias in recall is a common limit related to the majority of the studies carried out to investigate the parkinsonian personality 
. To the best of our knowledge, in fact, only one recent historical cohort study has been performed to investigate the possible premorbid personality in PD 
. At any rate, even if we cannot be sure that this condition was already present during the adolescence, due to the high prevalence of OCPeD found in the de novo
PD patients and in general in patients with a short disease duration as shown in , it could be probably considered an early manifestation of PD. Nowadays great attention is focused on the identification of possible markers of the early stage of the disease. Many non-motor symptoms occur early in PD and some of them, such as olfactory deficit, REM behaviour disorder, depression, constipation, may even predate also of many decades 
the diagnosis of PD which is based on motor signs 
. Nevertheless to confirm that PeDs could precede PD onset, prospective researches are required, even if defining the true premorbid time period could be impossible without valid biomarkers to reliably detect PD before the emergence of motor symptoms. Studies on a larger population of de novo
PD patients at early stage of disease are needed to support our findings.