Introduction

Parkinson’s disease (PD) is the most common neurodegenerative movement disorder. PD is associated with the progressive degeneration of dopaminergic neurons in substantia nigra pars compacta. The loss of dopaminergic neurons results in decreased levels of dopamine in the putamen of the dorsolateral striatum, producing dysfunctions of direct and indirect pathways of movement control that involve cortico-basal-thalamo-cortical loops [1]. Currently, PD treatment is symptomatic. The early symptoms of PD, such as rest tremor, bradykinesia and rigidity, are mainly treated with dopamine precursor compound levodopa. The oral combination of levodopa with peripheral dopa-decarboxylase inhibitor carbidopa provides an increase of levodopa availability in the central nervous system. To date, levodopa/carbidopa remains the gold standard of the treatment in PD [2]. However, long-term oral levodopa–carbidopa treatment induces motor and non-motor fluctuations and dyskinesias [3, 4]. These complications reflect the fluctuations in levodopa plasma concentrations that are associated with its short plasma half-life (~ 1.5 h) and erratic absorption resulting from delayed gastric emptying [5, 6]. In addition, several evidences support the hypothesis that pulsatile stimulation of striatal dopamine receptors contributes to the development of levodopa-induced motor fluctuations [7, 8].

Motor fluctuations together with non-motor symptoms, such as mood disorders, sleep disorders, cognitive dysfunctions and social cognition ability may impair patients’ quality of life (QoL) [9,10,11]. A recent study demonstrated that in some PD patients, the non-motor fluctuations cause a greater degree of disability and distress than the motor fluctuations [12].

In PD, the prevalence of affective disorders among patients’ caregivers is high and influences both burden and QoL [13]. Moreover, levodopa complications induce stress and burden in caregivers, leading to an impairment of their QoL [14].

Continuous dopaminergic stimulation with chronic infusion of levodopa reduces PD fluctuation complications [15, 16]. Levodopa/carbidopa intestinal gel (Duodopa®—Abbott Laboratories) is an aqueous gel that can be delivered to the jejunum via a percutaneous gastrojejunostomy tube (PEG-J tube) which is connected to a portable infusion pump dosing the drug gel continuously to the site of absorption. The delivery of Duodopa® directly in proximal small intestine provides a more stable plasma levodopa/carbidopa concentration and a continuous stimulation of the dopaminergic receptors in the striatum [17].

The most common adverse events of Duodopa® are related with the PEG procedures, such as abdominal pain, nausea, vomiting, infection and inflammation, and the infusion device, such as PEG dislocation, pump malfunction and tube occlusion. In addition, weight decrease and polyneuropathy have also been reported [18, 19].

Several clinical trials and observational studies demonstrated that Duodopa® reduces motor and non-motor fluctuations and dyskinesias, and improves the QoL of PD patients [20,21,22,23,24,25,26,27,28,29,30,31]. To our knowledge, no studies correlated among themselves the effects of Duodopa® therapy on motor complications, dyskinesias, anxiety, depression and QoL of PD patients. In addition, few studies have evaluated the Duodopa® therapy effects on caregivers’ QoL [32, 33].

The aims of this study were: (1) to evaluate Duodopa®-induced motor and non-motor outcomes (depression and anxiety, especially) and their impact on QoL in advanced PD patients; (2) to evaluate depression, anxiety and burden of caregivers and their relation to QoL before and after Duodopa® therapy of PD patients.

Materials and methods

Study population

In this open observational prospective study, 12 consecutive PD patients (mean age 73.83 ± 7.93; 7 males and 5 females), who started receiving treatment with Duodopa®, and their caregivers (mean age 57.58 ± 10.53; 6 males and 6 females) were enrolled. The clinical severity of PD patients was rated on the Hoehn–Yahr Scale (H&Y) and the Unified Parkinson’s Disease Rating Scale (UPDRS) [34]. All subjects gave written informed consent for participation in the study in accordance with the Declaration of Helsinki. IRCCS Centro Neurolesi “Bonino-Pulejo” Ethical Committee approved the study protocol.

Inclusion criteria were: diagnosis of PD according to UK Brain Bank criteria; motor fluctuations and dyskinesias refractory to conventional oral therapies; good levodopa responsiveness; normal cognitive functions. Exclusion criteria were: atypical Parkinsonisms; dementia according to diagnostic and statistical manual of mental disorders (DSM-V) criteria; concomitant neurological and psychiatric disease. The Table 1 shows the demographic and clinical characteristics of PD patients and their caregivers collected before treatment with Duodopa®.

Table 1 Socio-demographic and clinical characteristics of patients and caregivers
Full size table

Assessment

The assessments were completed before Duodopa® treatment initiation, at T0 (baseline visit), at T1 (3 months after Duodopa® treatment) and at T2 (6 months after Duodopa® treatment). At T0, for the screening of cognitive functions and behavioral disorders of PD patients, and then to exclude patients with cognitive impairment or neuropsychiatric disorders from the study, Montreal Cognitive Assessment (MOCA) [35] and Neuropsychiatric Inventory (NPI) [36] were used, respectively. At each visit, motor conditions and complications of PD patients were evaluated by means of UPDRS-Part III and IV (UPDRS-III and UPDRS-IV). Patients were assessed in “on-phase” for UPRDS III evaluation. The dyskinesias was evaluated by means of Unified Dyskinesia Rating Scale (UdysRS) [37]. Beck Depression Inventory (BDI-II) [38] and Hamilton Anxiety Rating Scale (HAM-A) [39] were used for evaluation of depressive and anxiety symptoms, respectively. Finally, QoL of PD patients was evaluated by means of Parkinson’s Disease Quality of Life Questionnaire (PDQ-39) [40]. These assessments were performed in “on state”.

The caregivers were evaluated at T0, T1 and T2 by means of BDI-II and HAM-A, for depressive and anxiety symptoms, respectively, Caregiver Burden Inventory (CBI) [41], for care burden, and SF-36 Health Status Questionnaire for QoL [42].

Before Duodopa® treatment, the results related to duration of the “off periods” are: nocturnal off-symptoms (twice-daily: 3.4 ± 3.7) and early morning off-symptoms (once-daily: 2.9 ± 2.7). After Duodopa® treatment, the results related to the duration of the “off periods” are: nocturnal off-symptoms (twice-daily: 2.4 ± 1.8) and early morning off-symptoms (once-daily: 1.9 ± 1.7).

Statistical analysis

A descriptive analysis of socio-demographic characteristics was performed and expressed as means and standard deviations for patient and caregiver groups. Normal distribution of data was evaluated using the Shapiro–Wilk normality test. The Wilcoxon signed-rank test was used to compare the results in three times (intra-groups analysis) for each group. To assess the magnitude of differences in scale scores in longitudinal comparisons, we calculated effect sizes by Cohen’s d [43]. Cohen classified effect sizes as small (d  =  0.2), medium (d  =  0.5), and large (d ≥ 0.8). According to Cohen, “a medium effect of 0.5 is visible to the naked eye of a careful observer”. A small effect of 0.2 is noticeably smaller than medium but not so small as to be trivial. A large effect of 0.8 is the same distance above the medium as small is below it [44]. Correlations among clinical variables (UPDRS-III and IV, UdysRS, BDI-II and HAM-A), total PDQ-39 and its sub-items for patient group, and clinical variables (BDI, HAM-A and CBI), two indices of SF-36 Health Status Questionnaire (ISM—Mental Health Status and ISF—Physical Status) and its sub-items for caregiver group were computed by Spearman’s coefficient. The analyses were performed using an open source R3.0 software package (R Foundation for Statistical Computer, Vienna, Austria). A 95% of confidence level was set with a 5% alpha error. Statistical significance was set at p < 0.05.

Results

Intra-group analysis

In the intra-group analysis for PD patients, we compared the scores of clinical scales, the scores of PDQ-39 and its sub-items between T0 (baseline) and T1, T1 and T2, T0 and T2. For caregivers, we compared the scores of BDI-II, HAM-A, CBI and the scores of two indices (ISM and ISF) of SF-36 Health Status Questionnaire and its sub-items between all the times.

In the patient group (Table 2), motor and non-motor symptoms as well as the QoL improved significantly between baseline and follow-up assessments. Indeed, we observed a significant decrease in scores of UPDRS-III, UPDRS-IV, BDI-II, HAM-A and PDQ-39 (including its sub-items) between all the times (p < 0.01) (Fig. 1). UdysRS scores did not show significant differences between T0 and T1, but showed significant differences between T1 and T2, and between T0 and T2.

Table 2 Intra-group analysis of patient group and assessment of the magnitude of differences in scale scores in longitudinal comparisons
Full size table
Fig. 1
figure 1

Trend of UPDRS-III, UPDRS-IV, UdysRS, BDI-II, HAM-A, PDQ-39 and its sub-items over time in patient group. All the differences are significant, excluding UdysRS scores between T0 and T2. BDI-II Beck Depression Inventory, HAM-A Hamilton Anxiety Rating Scale, PDQ-39 Parkinson’s Disease Quality of Life Questionnaire, UdysRS Unified Dyskinesia Rating Scale, UPDRS-III and UPDRS-IV Unified Parkinson’s Disease Rating Scale-Part III and IV

Full size image

In a similar way, in caregiver group (Table 3), we observed an improvement in anxiety, depression, level of care burden and QoL, as indicated by a significant decrease in scores of BDI-II, HAM-A, CBI and a significant increase in scores of SF-36 Health Status Questionnaire sub-items between all the times (p < 0.01) (Fig. 2). Only bodily pain sub-item scores did not change between T0 and T1. However, they increased at T2 although not reaching the level of significance (p = 0.17). In addition, role limitation due to physical health (p = 0.09) and general health perception (p = 0.37) sub-item scores did not show a significant difference between T1 and T2. Finally, ISF scores highlighted no significant difference, while ISM scores showed significant increase in every time point.

Table 3 Intra-group analysis of caregiver group and assessment of the magnitude of differences in scale scores in longitudinal comparisons
Full size table
Fig. 2
figure 2

Trend of BDI-II, HAM-A, CBI, two indices of SF-36 Health Status Questionnaire (ISM—Mental Health Status and ISF—Physical Status) and its sub-items over time in caregiver group. All the differences are significant, excluded ISF and BP scores between all the times, GH scores between T1 and T2, and PF between T0 and T2 and T1 and T2. BDI-II Beck Depression Inventory, CBI Caregiver Burden Inventory, HAM-A Hamilton Anxiety Rating Scale, ISF physical status, ISM mental health status. BP bodily pain, GH general health perceptions, MH mental health, PF physical functioning, RE role limitations due to emotional problems, RP role limitations due to physical health, SF social functioning, V vitality

Full size image

Correlation analysis of PD patients

At T0, the analysis reported no significant correlation between BDI-II and UPDRS-III scores (r = 0.4; p = 0.2), BDI-II and UPDRS-IV scores (r = − 0.35; p = 0.27), BDI-II and UdysRS scores (r = − 0.0008;p = 0.98), and between HAM-A and UPDRS-III scores (r = 0.27; p = 0.4), HAM-A and UPDRS-IV scores (r = − 0.17; p = 0.59), HAM-A and UdysRS scores (r = 0.19; p = 0.55). UPDRS-III, UPDRS-IV, UdysRS, BDI-II, and HAM-A scores did not show significant correlation with the scores of the total PDQ-39 and its sub-items. However, a significant trend for a positive correlation between PDQ-39 communication sub-item and HAM-A (r = 0.53; p = 0.08) was found (Table 4).

Table 4 Correlation among UPDRS-III, UPDRS-IV, UDyRS, BDI, and HAM-A scores and total PDQ-39 and its sub-item scores in patient group at T0 and T2
Full size table

At T2, there was no significant correlation between BDI-II and UPDRS-III scores (r = − 0.11; p = 0.37), BDI-II and UPDRS-IV scores (r = 0.23; p = 0.46), BDI-II and UdysRS scores (r = 0.49; p = 0.1), and between HAM-A and UPDRS-III scores (r = 0.21; p = 0.51), HAM-A and UPDRS-IV scores (r = − 0.13; p = 0.67), HAM-A and UdysRS scores (r = − 0.25; p = 0.43). BDI-II and HAM-A scores did not show significant correlation with the scores of the total PDQ-39 and its sub-items. A significant negative correlation between PDQ-39 emotional well-being sub-item and UPDRS-III (r = − 0.74; p = 0.006) and a significant trend for a positive correlation between PDQ-39 mobility sub-item and UPDRS-IV (r = 0.53; p = 0.07) and UdysRS (r = 0.54; p = 0.06) were found (Table 2).

Correlation analysis of caregivers

At T0, no significant correlation between BDI-II, HAM-A, CBI scores and ISF, ISM, and SF-36 Health Status Questionnaire sub-item scores was found in caregiver group (Table 3). At T2, no significant correlation between the scores of BDI-II and CBI and the scores of ISF and ISM, as well as between scores of BDI-II, HAM-A and CBI, and the scores of SF-36 Health Status Questionnaire sub-items were found. HAM-A scores showed a significant positive correlation with ISM (r = 0.56; p = 0.05) but not with ISF (r = − 0.05; p = 0.86) scores (Table 5). This demonstrates that the improvement of caregivers’ anxiety level improves the physical status domains but not the mental status domains.

Table 5 Correlation among BDI-II, HAM-A, and CBI scores and ISF, ISM, and SF-36 Health Status Questionnaire sub-item scores of caregiver group at T0 and T2
Full size table

Discussion

In this study, we evaluated on short time Duodopa®-induced changes of motor complications, dyskinesias, anxiety, depression and QoL of patients with advanced PD. As reported by other studies with short follow-up periods [18, 21,22,23, 27,28,29], 6 months after Duodopa® therapy, the motor complications and dyskinesias, assessed by UPDRS-III, UPDRS-IV and UdysRS, were significantly improved over time in all PD patients, as well as the depression and anxiety, as reported by BDI-II and HAM-A scores. At the same time, in agreement with other authors [19, 22, 23, 27, 28], PD patients’ QoL was significantly improved in respect to baseline conditions.

These results have encouraged us to assess the subsistence of possible correlations between the improvement of motor complications and dyskinesias, and the improvement of anxiety and depression of PD patients after Duodopa® therapy. In addition, we aimed to evaluate the possible cascading link between the reduction of these symptoms and the improved QoL. It is likely that the improvement motor fluctuations and dyskinesias of PD patients decreases their depression and anxiety and that the overall improvement of motor conditions and mood disorders may, in turn, improve the QoL.

At baseline, our results did not demonstrate a relationship among motor complications, dyskinesias, depression and anxiety in PD patients. Although anxiety and depression are associated with motor fluctuations and dyskinesias [45], these mood disorders have been commonly observed in patients with PD even independently of motor symptoms [46].

Although it is assumed that the motor and non-motor symptoms strongly influence how the patient perceives its QoL [47, 48], our results did not demonstrate a relationship among motor complications, dyskinesias, depression and anxiety, and QoL. It is likely that the small sample size does not provide the adequate statistical power to detect small correlations among all variables in a situation where even a moderate correlation trend may be an important outcome.

However, our study indicates that the improvement of patients’ communication after Duodopa® therapy is related to reduction of anxiety over time, supporting how the reduction of this symptom provides a greater control over its own autonomy in social relations.

In addition, as predictable, even after Duodopa® therapy, there was any relationship between the improvement of clinical conditions and mood disorders, and the improvement of overall QoL of PD patients. However, our results demonstrated that the improvement of motor conditions affected positively the emotional well-being perceived by patients, and the reduction of motor complications and dyskinesias had a moderately favorable impact on mobility sub-item of PDQ-39, supporting at least in part a relationship between some outcomes, whose improvement may be considered drug induced.

The different management of the therapy, as the continuous infusion of Levodopa, provides to patients more independence that, together with a greater adherence to therapy, could be responsible for the overall improvement of QoL, as demonstrated by others studies [49], even without its relation with clinical variables.

Overall, although the correlation results did not show significant relationship between clinical conditions and QoL of PD patients neither before or after therapy, this is the first study in which a possible relation between all clinical and psychological variables and QoL was assessed as consequence of drug effects.

The impact of PD on caregivers has been gradually recognized for high load of care, which requires a spending of physic, emotive and economic energy. This may result in anxiety and depression, which are present in the caregivers in all stages of the disease [50]. Furthermore, symptoms of anxiety and depression in caregivers may cause increase in burden and decrease of the QoL [51]. In our sample, at baseline, all caregivers showed anxiety, depression, burden of care and a poor QoL, especially in domains which concern social life, vitality, mental health; moreover, even the symptoms related to the emotional sphere were disabling. Six months after Duodopa® therapy, all caregivers showed a significant improvement in anxiety, depression, level of care burden and QoL, as assessed by BDI-II, HAM-A, CBI and of SF-36 Health Status Questionnaire, respectively. However, only mental status domains of SF-36 Health Status Questionnaire improved over time, while physical status domains remained unchanged, suggesting a possible improvement of emotional aspects only. The overall improvement in caregivers could be explained with the mobility improvement: the patient regains their own autonomy easing the caregivers in the physical care. Consequently, it is likely that even depression, anxiety and QoL of caregivers improve.

In the present study, we did not find evidence of relationship between the levels of anxiety, depression, and burden and QoL of caregivers neither before nor after therapy of PD patients. However, a reduction of the anxiety after therapy correlated with mental status domains of SF-36 Health Status Questionnaire, supporting the greater involvement of caregivers’ emotional spheres and psychological status in the patients’ care.

In conclusion, Duodopa® therapy improves short-term motor and non-motor conditions of patients with advanced PD and their QoL. However, motor fluctuations, dyskinesias, anxiety, depression and QoL have not significant mutual effects. At the same time, in caregiver group, anxiety, depression, burden and QoL improve after therapy, but there is no significant relationship among them. A further study with a larger population and long-term follow-up could allow us to confirm these results or not. Moreover, a future research on a larger population could be a direct assessment of impact of Duodopa®-induced motor and non-motor outcomes on psychological status, burden, and QoL of caregivers.