Introduction

Lewy body dementia (LBD) is the second most common neurodegenerative dementia, comprises both dementia with Lewy bodies (DLB) and Parkinson’s disease dementia (PDD) [1]. Though the two diseases are distinguished clinically from “1-year rule” (in PDD the motor symptoms precede the onset of dementia by at least 1 year) [2], their association remains to be clarified because of pathological and genetic overlap on Lewy body disease continuum [3].

LBD patients present with a wide range of motor and nonmotor (cognitive, neuropsychiatric, sleep, autonomic nervous dysfunction) symptoms. Numerous studies have paid attention to “geriatric syndromes” of dementia older adults, which can increase the disability and dependence [4], play an important role in dementia diagnoses and management as well. Previous studies point that polypharmacy [5], falls [6], fear of falling (FoF) [7], frailty [8], orthostatic hypotension (OH) [9], urinary and fecal incontinence [10], depression [11], and sleep distribution [2] may increase cognitive impairment, neuropsychiatric symptoms, and dependence on daily living, even in early-stage dementia subtypes [12]. What’s more, APOE ε4 allele is a strong risk factor across the Lewy body disease spectrum, and it has been confirmed to be associated with the occurrence of LBD, and can accelerate LBD progression and cause poor prognosis [

  • OH: Over 20 mmHg decrease in systolic blood pressure or > 10 mmHg decrease in diastolic blood pressure within 3 min of standing or have symptoms of postural dizziness.

  • Constipation: Defecation less than three times a week, hard stool, having to strain to pass stools.

  • Urinary symptoms: Complaining of frequent or urgent micturition, dysuria, or urinary incontinence, except for urinary tract infection.

  • HH: Suffering from excessive sweating required to regulate normal body temperature during the day or at night.

  • Falls: Having fallen in the previous year except for trip** on a rug and slip** on wet floor.

  • Fear of falling: Fearing of falling in daily activities and having The Falls Efficacy Scale International > 16.

  • Delusion: The caregivers responded “Yes” to items 1 on the NPI.

  • Depression: The caregivers responded “Yes” to items 4 on the NPI.

  • Anxiety: The caregivers responded “Yes” to items 5 on the NPI.

  • Apathy: The caregivers responded “Yes” to items 7 on the NPI.

  • Loss of appetite: The caregivers reported that the patient had a loss of appetite and a lack of desire to eat, then responded “Yes” to items 12–1 on the NPI.

  • Insomnia: Insomnia Severity Index score ≥ 8.

  • Excessive daytime sleepiness: Epworth Sleepiness Scale score ≥ 11.

    • APOE genoty**

    We used peripheral blood to extract genomic DNA, and polymerase chain reaction was amplified to determine the APOE gene in this study. All the detailed method had been described at our previous study [27]. We were sure about all genotypes without knowledge of the patient status.

    Statistical analyses

    Descriptive analyses were conducted using frequency for qualitative variables and mean [± standard deviation (SD)] or median [interquartile range (IQR)] for quantitative variables. Student’s t-test was run to compare the two independent groups (DLB and PDD) for normally distributed data and a Mann-Whitney U test was used for nonparametric data. The comparisons among the four groups [DLB with APOE ε4 (+), DLB with APOE ε4 (−), PDD with APOE ε4 (+), PDD with APOE ε4 (−)] were conducted by Mann–Whitney U tests and Bonferroni corrections were applied. Qualitative variables were assessed using a chi-squared test. Analysis of correlation was made by using Pearson and Partial (adjusted gender, education, habits of smoking and alcohol consumption, heart disease, hypertension, diabetes mellitus, stroke and CDR) correlations to evaluate the correlation between APOE ε4 allele, basic and clinical characteristics in probable DLB and PDD. Logistic regression models (adjusted gender, education, habits of smoking and alcohol consumption, heart disease, hypertension, diabetes mellitus, stroke and CDR) were used to analyze the associations between APOE ε4 allele and geriatric symptoms, and the results were recorded with odds ratios (ORs) and 95% confidential interval (95% CI).

    For the statistical analyses, the IBM Statistical Package for the Social Sciences (SPSS) for Windows (version 22.0; IBM Corporation, Armonk, NY) was used. Adjusted odds ratios (OR) are presented with 95% confidence interval (95% CI). A p - value of less than 0.05 was considered significant. All tests were performed bilaterally.

    Results

    Clinical characteristics

    A total of 185 patients, 92 with mild-moderate DLB and 93 with mild-moderate PDD, were included in the study (Table 1). The median and mean ages of DLB patients, both age at visit and onset, were older, and DLB patients had higher proportions of hypertension (32.6% vs. 17.2%, p = 0.015) and DM (23.9% vs. 10.8%, p = 0.018), as well as were more APOE ε4 carriers (57.6% vs. 12.9%, p < 0.001) than PDD patients. Comparing to patients with PDD, the DLB cases displayed more frequency of fluctuating cognition (28.3% vs. 9.7%, p = 0.001), VH (47.8% vs. 14.0%, p < 0.001) and RBD (55.4% vs. 22.6%, p < 0.001) at the time of diagnosis, while there was no significant difference in APOE ɛ4 allele subgroups (Additional file, Table S1).

    Table 1 Sample characteristics
    Full size table

    The global and sub-item scores of MMSE and MoCA were significantly higher in the PDD cases than DLB cases at their first visit (Table 2). And patients with PDD had significantly greater H&Y stage (mean score: 2.12 ± 0.64 vs. 0.98 ± 1.50, p = 0.000) higher UPDRS part III scores (mean score: 30.76 ± 14.48 vs. 7.80 ± 1.50, p = 0.000). While the patients with DLB had higher scores of NPI (mean score: 15.14 ± 13.77 vs. 2.65 ± 4.49, p = 0.000) and ADL (mean score: 32.77 ± 12.74 vs. 26.04 ± 10.15, p = 0.000). We also found that the presence of APOE ε4 allele was not significantly correlated with the age at visit, the age at onset, scores of MDS-UPDRS III, H&Y stage, ADL, MMSE, MOCA and NPI, as well as the presences of fluctuating cognition, VH, parkinsonism and RBD in both groups, except for the scores of MoCA-language (r = − 0.264, p = 0.016) in DLB cases and MMSE-registration (r = − 0.216, p = 0.038) in PDD cases (Additional file, Table S2). Besides, there were no significant associations between clinical core features of LBD and APOE ε4 allele after adjusting confounders (Additional file, Table S3).

    Table 2 Cognitive and motor function in patients with probable DLB and PDD
    Full size table

    Associations between APOE ε4 and geriatric symptom complaints in DLB and PDD

    In patients with DLB, the most frequent geriatric symptom complaints were anxiety (50.0%), followed by depression (48.9%), apathy (45.7%), loss of appetite (38.0%) and delusion (34.8%), while constipation (53.8%), insomnia (49.5%), FoF (45.2%), OH (25.8%) and falls (25.8%) troubled patients with PDD mostly, as summarized in Table 3. The presence of constipation, FoF, and insomnia were more common in the PDD than DLB (p < 0.05), and delusion, depression, anxiety, apathy, and loss of appetite were more common in the DLB than PDD (p < 0.001). And the patients with DLB + APOE ε4 (+) were presented more delusions (p = 0.005) and apathy (p = 0.007) than patients with PDD + APOE ε4 (+). But there were no significant differences of geriatric symptoms between APOE ε4 carriers and APOE ε4 non-carriers in both DLB and PDD groups.

    Table 3 Comparation of geriatric symptom complaints between probable DLB and PDD
    Full size table

    Logistic regression models in Table 4 showed that patients APOE ε4 allele was associated with the presences of HH (OR = 3.472, 95%CI: 1.082–11.144, p = 0.036) and depression (OR = 3.002, 95%CI: 1.079–8.353, p = 0.035) in the mild – moderate DLB patients after adjusted gender, age at visit, education, habits of smoking and alcohol consumption, heart disease, hypertension, diabetes mellitus, and stroke, while there was no associations between APOE ε4 allele and geriatric symptoms in mild-moderate PDD patients.

    Table 4 Associations between APOE ε4 allele and geriatric symptom complaints in DLB and PDD
    Full size table

    Discussion

    Our study reveals that the geriatric symptoms, particular constipation, FoF, delusion, depression, anxiety, apathy, insomnia, loss of appetite, are highly prevalent in mild-moderate LBD patients. The APOE ε4 allele was highly significant over represented and associated with HH and depression in DLB patients. While APOE ε4 allele was not associated with global cognitive impairment, motor function, activities of daily life, as well as four clinical core features in mild-moderate LBD patients.

    Despite APOE ε4 was reported to be risk factors for LBD (including DLB and PDD) in previous literature [28], the frequency of APOE ε4 carriers in DLB and PDD was not sure. The APOE ε4 carriers was about 30 - 60% in DLB [29, 30] and about 30–40% in PDD [31, 32], but there was not significantly difference between them [32]. Our data showed the frequency of APOE ε4 carriers in DLB was line with previous studies, while higher than in PDD.

    In current study, we found that 28.3% DLB patients had fluctuating cognition, 47.8% had VH, and 55.4% experienced RBD, which were less prevalent than our previous research [33]. We analyzed the percentages of the core features among 37 probable DLB patients and reported that 64.9% displayed fluctuation, and 73.0% showed VH, the difference was possibly due to the mild-moderate stage of these samples. While a longitudinal prospective cohort study with 100 patients with DLB, showed that about 39% patients had VH, 46% manifested fluctuating cognition, and 77% patients experienced RBD in early stage [34]. Moreover, we didn’t find significant associations between APOE ε4 allele and core clinical features in LBD patients. APOE ε4 allele is reported to be associated with hallucinations and Alzheimer’s disease (AD)-pathological protein in parkinsonian syndromes [35, 36], patients with DLB + APOE ɛ4 showed more pronounced hallucinations in atypical Parkinson’s disease (PD) cohort [37]. Consistent with our research, Gan-Or et al. [38] reported a lack of association between the APOE ε4 allele and risk of RBD in LBD patients, and this finding was confirmed by Sunwoo et al. [39], who demonstrated no difference of APOE genotypes in patients with RBD as well as general population. And APOE ε4 allele was not associated with the incidental parkinsonism symptoms in parkinsonian syndromes, it could only derive progressive cognitive impairment but not motor progression in parkinsonian syndromes [40, 41].

    We also found higher proportions of constipation, FoF, and loss of appetite in PDD, while the insomnia and NPS, such as hallucinations, delusions, anxiety, depression and apathy, were more frequent in DLB patients. Constipation is the most prominent gastrointestinal manifestation in patients with PDD and DLB, and it may cause by neuronal loss in the myenteric plexus [42]. The enteric nervous system (ENS) origin suggested that the increased intestinal permeability in PD triggered alpha-synuclein aggregation within the ENS [43], so as to affect the gastrointestinal emptying function and ultimately produced PD [44]. Gastrointestinal dysautonomia also played a possible role in the etiology of constipation in PD, which had been proved in dopamine transporter single-photon emission computed tomography imaging in Hinkle‘s research [45]. Constipation increased the risk of develo** PD [46] and noting that individuals experiencing the constipation may suffered from motor symptoms of PD with a mean of 18.7 years [47]. At the same time, DLB and PDD patients also have a certain proportion of gastroparesis [48]. Gastroparesis presented as nausea, reduced appetite, early satiety, bloating, vomiting, and weight loss [49], about 25–45% PD patients reported subjective symptoms [50]. Loss of appetite may be caused by prolonged gastric emptiness, which has an adverse interaction with constipation. The NPS in mild - moderate DLB are more frequent than PDD, which is consistent with many studies [51]. Depression is particularly common in DLB and established as a supportive feature in the revised criteria for the clinical diagnoses [2], and the high depression rates assessed by NPI are also reported in several cohort patients with mild dementia [37, 52]. A cohort with 223 dementia cases [52] (56 with DLB and 11 with PDD) reported significantly more depression in DLB but there was no relationship between the presence of APOE ε4 allele and depression. However, our study revealed opposite results. The APOE ɛ4 allele was associated with depression in mild-moderate DLB after adjusted several confounders. Several studies had reported an association between APOE ε4 and depressive phenomenology [53, 54]. APOE ε4 allele can disturbed the clearness of amyloid plaques stored in limbic regions to regulate emotional processes, and the deficits if cholinergic [55] and noradrenergic [56] activities in APOE ε4 carriers might explain the more severe depressive symptoms in mild DLB.

    In addition, there are few studies comparing the falls and FoF of DLB and PDD. In a prospective study involving 30 DLB and 40 PDD patients, the incidence of falls (86.8%) was higher in PDD than in DLB (69.2%) [6]. Patients with DLB sustained 6 times, and PDD sustained 20 times more falls than normal cognition older adults [57]. The old age, sleepiness, dementia, autonomic nervous dysfunction, and freezing of gait were all risk factors of falls [58]. However, the research on FoF is mostly limited to DLB and AD. In a comparative research among old adults with DLB, AD, and non-dementias, DLB patients had much higher proportion of FoF (86.9%) than those with AD (36.0%) and without dementia (37.4%) [59]. Parkinsonism, visual dysfunction (including visual hallucination, color vision impairment, visuospatial disorder, and decreased occipital lobe activity), OH, and attention disorders may provide plausible pathways causing development of FoF [12]. Patients with PDD had more proportion of insomnia in our findings, which could be proved in previous research. PDD patients slept shorter, had more night-time behavior disturbances, and were less satisfied with their sleep than patients with DLB [11], which could prove our finding. Restless legs syndrome (RLS), turn-over difficulty and pain [60] were in a higher proportion, which may be possible reasons for insomnia in PDD patients. The loss of hypocretinergic neurons in the hypothalamus could decrease hypocretin and disturb sleep-wake cycle regulation, resulted to insomnia in PDD [61]. What’s more, dopaminergic dysfunction with impairment of medial thalamic pain system in PD could cause RLS, and the associations between RLS and insomnia in patients with LBD remained to explore.

    Patients with DLB demonstrated a poorer cognitive situation at their visit than patients with PDD. The ages at onset and at visit were both older in DLB than PDD, which was consistent with our multicenter findings [62] in Chinese memory clinics. Moreover, patients with DLB suffered more from hypertension and DM than PDD patients. Take as the fact that the older ages at onset and visit as well as the higher mean score of CDR in DLB into consideration, once again the possibility that there may be the demographic factors driving this cannot be discounted.

    We provide the relatively large sample size and reliable information from outpatient records in current clinical research on DLB and PDD in early stage, while we recognize some limitations in this study. Firstly, the lack of specificity biomarkers (even pathological confirmation) and objective polysomnography could lead to bias in the determination of disease and RBD. And since all patients came from a single center, there existed a misclassification bias to some extents. Some relevant factors may be not taken into account in this research, which need further consideration in future.

    Conclusion

    The presence and co-incidence of geriatric symptoms, including sleep disorders, autonomic nervous dysfunction, and NPS are common in patients with mild-moderate DLB or PDD. The APOE ε4 allele was represented in DLB, and is associated with hyperhidrosis and depression, but not global cognition, activitives of daily life, motor function and other neuropsychitric symptoms. While APOE ε4 allele plays no role in the clincal characteristics of PDD. These findings can signifiant improve the awareness of geriatric symptoms, and contribute to the healthcare management of mild-moderate DLB and PDD.