Cohort study on somatoform disorders in Parkinson disease and dementia with Lewy bodies

Recent literature on the psychiatric manifestations of Parkinson disease (PD)¹ identified depression and impulse control disorders as mental disorders preceding or occurring concomitantly with parkinsonism and its treatment. Yet, “other mental symptoms” need to be “recognized as sources of disability for patients with PD”²: a recent survey on patients hospitalized for PD³ found that prior hospitalizations for mental disorders other than depression equally increased the risk of PD occurrence. The present study hypothesized that somatoform disorders (SFMD) might be part of the presentation of PD and dementia with Lewy Bodies (DLB).

SFMD is the general category grouping, on the basis of medically unexplained symptoms, somatization and conversion disorders (once defined as hysteria), pain, and hypochondria, while delusional disorders somatic type identifies a bordering category.⁴ Several reasons anchored our interest on SFMD in PD: SFMD might precede the onset of neurodegenerative disorders,^5,6 conversion type SFMD were described anecdotally in patients who developed PD^7–9, alexithymia (inability to distinguish feelings from bodily sensation of emotional arousal) is increased in PD,¹⁰ and, finally, the remarkable intensity of response to placebo in PD^11,12 evidences that there is suggestibility in PD and suggestibility was historically defined as the hallmark of hysteria-SFMD.^9,13 We therefore designed a cohort study based on direct observation of SFMD symptoms and semistructured interviews, investigating hospital and general practitioners' (GP) reports, describing 1) cohort prevalence of SFMD in different neurodegenerative diseases; 2) presentation and phenomenology of SFMD symptoms; and 3) comparisons between patients with PD and patients with DLB with and without SFMD, at onset and during disease progression.

METHODS

This prospective study was designed following recommendation of the STROBE statement.¹⁴ It was based on a cohort of 1,360 new patients evaluated from 1999 to 2008 in our movement disorder and memory outpatient clinics in Central Italy, Adriatic Coast, serving a population of 1,300,000. Recruitment began in 2001 and ended in 2004 for PD and DLB, and in 2006 for the other neurodegenerative diseases. Two cases, initially considered incidental SFMD, were observed in 1999. As a comparator prevalence estimate from the general population of the same area referral system, we accessed patients with SFMD from our psychiatry clinic, following 1,718 patients from 1999 to 2006.

Categorization of patients into SFMD and non-SFMD groups was based on direct observation of symptoms in the year (6 months before/after) coincident with the diagnosis of neurodegenerative diseases. As a second assessment, all patients underwent semi-structured interviews by a rater blinded to neurologic diagnoses. Interviews, based on DSM-IV-TR manual,⁴ investigated somatic complaints with examples and a checklist presented to patients and caregivers, focusing on SFMD traits, such as dependency, mannerism, viscosity, adoption of a sick role, and histrionic dramatic descriptions of disease (e-Methods on the Neurology® Web site at www.neurology.org). Evaluations of past SFMD symptoms included information from prior hospital records and reports from each patient's GPs over the past 4–20 years. Insight, somatic type delusional disorders were specifically investigated as part of the Neuropsychiatric Inventory (NPI) assessment.¹⁵

The PD diagnosis was made according to the UK Parkinson's Disease Society Brain Bank criteria.¹⁶ Parkinsonism was evaluated with Unified Parkinson's Disease Rating Scale (UPDRS)¹⁷ and Hoehn & Yahr (H&Y) scale. Nonmotor symptoms of PD (olfactory and visual disorders, constipation, inner tremor, pain)² were analyzed as possible confounding factors. Diagnosis of probable DLB was based on Consensus Criteria.¹⁸ Probable Alzheimer disease (AD) diagnosis was based on National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders Association criteria.¹⁹ Progressive supranuclear palsy (PSP) diagnosis was based on National Institute of Neurological Disorders–Society for Progressive Supranuclear Palsy criteria.²⁰ Multiple system atrophy (MSA) diagnosis was based on Consensus Statement.²¹ Frontotemporal dementia (FTD) diagnosis was based on standardized clinical criteria.²² Neuroimaging and further diagnostic evaluations are reported in e-Methods. All patients with PD, DLB, MSA, or PSP underwent l-Dopa challenges.²³ There were no treatment restrictions for inclusion in the study.

All patients underwent Mini-Mental State Examination (MMSE),²⁴ Frontal Assessment Battery (FAB),²⁵ NPI,¹⁵ and the SFMD Semi-Structured Interview, and were followed for 2 years in order to confirm diagnosis.

Two comparison studies were performed between patients with PD and patients with DLB with and without SFMD.

The first compared 29 PD-SFMD and 15 DLB-SFMD patients at initial diagnosis with 87 patients with PD and 45 patients with DLB without SFMD, randomly selected and matched for age (within 3 years) and disease duration (6 months). These patients were also evaluated with the Clinician Assessment of Fluctuations (CAF),²⁶ the Beck Depression Inventory (BDI),²⁷ the SCL-90R, somatoform disorder subitem,²⁸ and polysomnography for REM sleep behavior disorder. All patients with PD with impulse control disorders or with incongruent neuroimaging studies or definite genetic mutations were excluded from possible SFMD evaluation and from inclusion in the matched PD group.

The second comparison was a 4-year follow-up assessment of motor and cognitive outcomes. A total of 66 patients with PD and 30 patients with DLB without SFMD, from the initial comparative group, were matched with SFMD patients completing follow-up, for educational level (±2 years) and (MMSE) score (±1 point), at onset of study. The 4-year follow-up study included MMSE, Dementia Rating Scale–2 (DRS-2),²⁹ and UPDRS total assessments.

Institutionalization rates (admission to long-term care facility or need for full-time caregiver employment) were assessed in the patients with PD and patients with DLB at 9 years after study enrollment. Figure e-1 shows study design.

RESULTS

Comparative studies.

Comparisons at initial diagnosis of PD or DLB between patients with or without SFMD matched for age and disease duration showed no differences for patients with DLB and only minor differences for patients with PD. University education was significantly more frequent in PD-SFMD patients. Gender, neuropsychological test scores, and l-dopa equivalent doses were similar³² (table 3).

View this table:

• View inline
• View popup
• Download powerpoint

Table 3 Demographic, clinical, neuropsychological, and neuroimaging data at admission to the study: comparisons of SFMD patients with age and disease duration–matched patients with PD and patients with DLB

SCL-90R somatic disorder subitem and NPI scores were higher in PD-SFMD patients compared to patients with PD without SFMD (p < 0.05), but no differences were observed in other SCL-90R subitems. Depression was equally present in patients with and without SFMD as shown by BDI scores. The comparison with all newly diagnosed patients with PD (n = 236 without impulse control disorders or genetic mutations) showed that PD-SFMD patients were older (70.6 ± 5.5 vs 63.9 ± 9.5, p < 0.05). No feature distinguished the presence from absence of SFMD in patients with DLB. MMSE, DRS-2, and FAB scores, frequency of RBD and visual hallucinations, and treatment doses were all statistically different (p < 0.05–0.01) in the comparison DLB vs PD groups (with and without SFMD) at baseline. Comparisons during and at the end of the 4-year follow-up between patients matched for age, disease duration, educational level, and MMSE score at initial diagnosis showed instead relevant statistical differences (table 4).

View this table:

• View inline
• View popup
• Download powerpoint

Table 4 Demographic, clinical, and neuropsychological data at admission to the study and after 2 and 4 years of follow-up in SFMD patients and patients with PD or DLB matched for education and MMSE score at baseline

Rate and slope of MMSE and DRS-2 changes through time separated PD-SFMD and both DLB groups from PD without SFMD (p < 0.001 for every comparison) at each evaluation (p < 0.001). Regression analysis showed that MMSE and DRS-2 of PD-SFMD and DLB patients with or without SFMD could be described by a linear fit from baseline to 4 years follow-up, thus showing that progression of cognitive decline (figure 1) was similar in DLB and PD-SFMD patients and different from PD without SFMD. Controlling for age and education, the PD-SFMD, as compared to PD only patients, showed a significantly lower reduction in DRS-2 score over time (regression coefficient for PD only patients = −8.16; 95% confidence interval −7.55 to −8.76; p < 0.001), confirming the large difference observed in the univariate analysis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Comparison of Dementia Rating Scale–2 (DRS-2) slopes of progression in the 4 groups of patients of the comparative study

The progression of cognitive decline in Parkinson disease–somatoform disorder (PD-SFMD) and in dementia with Lewy bodies (DLB) was similar, as evidenced by similar decrements from the 2-year to the 4-year assessments in PD-SFMD and from the admission to the 2-year assessments in DLB.

Progressive worsening of NPI scores was observed in all groups, but significant differences distinguished PD-SFMD from PD without SFMD and both DLB groups, throughout follow-up. Further assessments evidenced the progression of cognitive and behavioral decline in SFMD patients. Cognitive fluctuations appeared in 9 PD-SFMD patients and, at 4 years, CAF score separated PD-SFMD from PD group (p < 0.001, table 4), while at baseline no differences had been observed between the 2 groups. VH were only present in DLB at baseline (p < 0.001), while 13 PD-SFMD patients had VH at the end of study. Only 2 patients with PD without SFMD reported VH at follow-up (table 4, p < 0.001). At the end of the 9-year follow-up, all PD-SFMD and DLB patients were institutionalized, whereas none of the patients with PD required institutional care.

DISCUSSION

Previous descriptions of SFMD^7–9 and catatonic signs^33,34 in PD are rare, and to our knowledge, there are no reports presented on DLB. This report shows that SFMD was identified in 7% of patients with PD and 12% of patients with DLB at the time of primary neurologic diagnosis and can recur during follow-up. SFMD symptoms were more frequent in PD and DLB than in AD, FTD, MSA, and PSP, and more frequent than in our psychiatric population. The rates in PD and DLB were higher than rates reported for conversion or somatization disorders in the general population.⁴

The design of the study, with patients included only if actual SFMD signs were observed, with verification through interviews (assessing psychiatric traits), with hospital and GP reports, and with follow-up evaluations, allowed us to assess the consistency of SFMD patterns during or before neurologic disease course, and showed features of SFMD in PD and DLB which were different from features reported in DSM-IV-TR for conversion or somatization disorders.⁴

Educational levels, gender, and age of patients with PD and patients with DLB were different from our psychiatric population and from literature.⁴ The SFMD phenomenology was also partly different because SFMD in PD and DLB tended to recur, and to shift among motor, sensory, and pain syndromes. Symptoms were often characterized by delusional qualities (body deformation, requests for invasive medical procedures), and motor catatonic signs were often observed (table 2, table e-1, table e-2). Therefore, the general category of SFMD, rather than DSM-IV-TR subcategories, was purposely used to encompass variant presentations.

We acknowledge that we may have underestimated the prevalence of SFMD, as PD-DLB patients whose interviews were not matched by observation of SFMD symptoms were excluded: recently high somatization SCL-90R scores were found in 40% of patients with PD,³⁵ suggesting higher prevalence than in our study. Nevertheless, obtaining consistency of SFMD patterns was our intention. The main conclusion of our study highlights not just the prevalence but also the relevance and impact of SFMD on disease progression. We showed that identification of SFMD in patients with PD has prognostic relevance: SFMD had higher prevalence in patients with DLB than in patients with PD, and while no differences were observed initially between patients with PD with or without SFMD, all PD-SFMD patients were eventually diagnosed with PD with dementia (PDD); slopes of progression of DRS-2, FAB, and MMSE were similar in PD-SFMD and in DLB patients; and cognitive fluctuations and visual hallucinations appeared in DLB and PD-SFMD. All findings suggest that SFMD in patients with PD is a predictor of the development of dementia of the DLB type.

The relationship of SFMD with dementia in DLB and PD has practical implications.

First, in PD, identification of SFMD might be useful to expand selection criteria for inclusion/exclusion of patients in therapeutic procedures (e.g., transplant, electrode implants) as inclusion of patients who predictably develop dementia would interfere with efficacy/safety evaluations. Second, SFMD with variable patterns, including conversion, or somatic delusion and catatonic signs might be useful in the diagnosis of DLB or PDD, or might be added to definitions of clinical features in DLB and PDD.¹⁸

Thus, the main message of our study is that complex somatization patterns can be observed and should be searched in PD and DLB, as their identification can predict cognitive outcome, or support diagnoses. Recent neurobiological evidence suggests that hysteria (conversion-somatization SFMD⁴) and catatonia, both characterized by overactivity of orbitofrontal and anterior cingulate gyrus,^36–38 might both represent the expression of an archetypal sham death reflex (playing possum) present in animals as a life-saving behavior^39,40 and could thus be considered a continuum. Indeed, our study shows that hysteria-catatonia symptoms might appear as a continuum in PDD and DLB, and suggests thus that investigating their mechanisms in these neurodegenerative diseases might help clarify their features in other conditions.

AUTHOR CONTRIBUTIONS

Statistical analysis was conducted by Dr. Lamberto Manzoli.

ACKNOWLEDGMENT

The authors thank Prof. C.G. Goetz for editing the manuscript and addressing the presentation and Prof. P. Calabresi, A. Antonini, A. Padovani, P. Barone, G. Wenning, F. Stocchi, S. Jonas, and Dr. P. Tiraboschi, F. Gambi, and G. Sepede for reading and discussing the manuscript.

DISCLOSURE

Dr. Onofrj served on scientific advisory boards for GlaxoSmithKline, Novartis, Lundbeck, Inc., Eisai Inc., Valeant Pharmaceuticals International, Medtronic, Inc., and Newron; and receives research support from the Italian Ministry of Health and Italian Ministry of Education. Dr. Bonanni has received research support from the Italian Ministry of Health. Dr. Manzoli has served as a consultant for Pfizer Inc. and has received research support from the Italian Ministry of Health. Dr. Thomas serves on a scientific advisory board for UCB; has received funding for travel from GlaxoSmithKline; and has received honoraria from Boehringer Ingelheim and GlaxoSmithKline.