Clinical characteristics of fatigued Parkinson’s patients and the response to dopaminergic treatment

This was an observational cross-sectional study. Two hundred and twenty-two PD patients, diagnosed according to the UK PD Brain Bank criteria [14], were recruited between December 2012 and April 2014 from the outpatient Department of Neurology of the First Affiliated Hospital of China Medical University (Shenyang, China). Patients with other forms of parkinsonism were excluded. Patients with dementia or fatigue related to a disease other than PD were excluded (e.g., peripheral circulatory disorders, such as ischemic heart disease and left ventricular failure, or acute and chronic inflammatory demyelinating polyneuropathies, such as Guillain-Barre syndrome). All patients enrolled in the study gave written informed consent. The Ethics Review Board of First Affiliated Hospital of China Medical University (Shenyang, China) approved the protocol of the study. Demographic data were collected for all PD patients. The total amount of dopaminergic medication was expressed as the levodopa equivalent daily dosage (LEDD), which was determined using methods previously reported in a systematic review of LEDD [15].

Disease severity was evaluated during the on phase using the Unified Parkinson’s Disease Rating Scale (UPDRS) and Hoehn & Yahr staging (H-Y stage) [16]. As previously described [17], the following sub-scores obtained from specific items of UPDRS III were calculated as following: tremor (items 20 and 21), rigidity (item 22), bradykinesia (items 23–26 and 31) and posture/gait (items 27–30).

Fatigue was assessed using the Parkinson Fatigue Scale-16 (PFS-16). The PFS-16 is a 16-item self-report scale that is specifically designed to assess the physical aspects of fatigue in PD patients [1]. The item response options range from 1 (“strongly disagree”) to 5 (“strongly agree”). The total PFS-16 score ranges from 1 to 5 and is obtained by dividing the sum of all item scores by 16 [18]. A cut-off score of 3.3 was used to indicate the presence of fatigue [1]. Subjects with a PFS-16 score <3.3 were classified as nonfatigued, and those scoring ≥3.3 were classified as fatigued.

The Montreal Cognitive Assessment (MoCA) Beijing version was used to evaluate cognitive functioning. Because depression and sleep disturbances have been reported to affect fatigue, we revised our study protocol during the study. The 17-item Hamilton Depression Scale [19] (HAMD) and Parkinson’s Disease Sleep Scale (PDSS) [20] were administered to 177 of the 222 patients to assess depression and sleep disturbance, respectively. In the 17-item HAMD, a higher score indicates severe depression, and depression was classified as follows: no depression (0–7); mild depression (8–16); moderate depression (17–23); and severe depression (≥24) [21]. The range of PDSS score was 0–150, and a lower score indicated a higher number of sleep disturbances.

All continuous data are presented as the mean ± standard deviation (SD). Categorical variables were compared with chi-square tests; the Mann-Whitney U test was used for continuous variables. Correlations were assessed using Spearman rank order correlation coefficients. The odds ratios were calculated using logistic regression and chi-square tests. All data analyses were performed using SPSS software. A value of P < 0.05 was considered statistically significant.

According to the suggested cut-off score for the PFS-16 (≥3.3), fatigue was present in 132 (59.46 %) patients. Comparisons of the demographic and clinical characteristics between the fatigued and nonfatigued groups are listed in Table 1. The mean disease duration in the nonfatigued group was significantly shorter than that in the fatigued group (3.75 ± 3.26 years vs. 4.61 ± 2.99 years, p = 0.046). There were no significant differences between the two groups in other clinical characteristics, including sex, age, onset age and education.

Disease severity, as determined by H-Y stage and the total UPDRS score and all subscales, was more serious in the fatigued group than in the nonfatigued group (Table 1). Further analysis of the individual motor symptoms revealed that except for tremor, most motor symptoms, including rigidity, bradykinesia and posture/gait, were more serious (higher scores) in the fatigued group than in the nonfatigued group (Table 1). Similarly, the fatigued group had significantly worse scores for depressive symptoms and sleep disturbance than the nonfatigued group had (Table 1).

The LEDD and MoCA scores tended to be higher in the fatigued group than in the nonfatigued group, but this was not significant.

As shown in Table 2, the univariate logistic regression revealed that longer disease duration, a greater disease severity (H-Y stage, total UPDRS score and all subscales), depression and sleep disturbances were associated with fatigue (p< 0.05). Other clinical variables, including gender, age onset age, education, LEDD and MoCA score were not significantly associated with fatigue.

We further explored the independent risk factors of fatigue in PD patients using multivariate logistic regression. Only sleep disturbance (odds ratio = 0.974, 95%CI 0.959–0.989, P = 0.001) remained an independent risk factor for fatigue (Table 2). Disease duration, disease severity (H-Y stage, total UPDRS score and all subscales) and depression were not associated with fatigue (Table 2).

To further explore the clinical profiles of dopaminergic drug responsive fatigue, we examined the data of 79 patients in the fatigued group with complaints of fatigue and who were treated with dopaminergic drugs. We defined the fatigue amelioration rate between 0 % and 100 %, and assigned fatigue remission rate into 4 level (level 0: no remission, remission rate <25 %; level 1: mild remission, remission rate between 25 % and 50 %; level 2: moderate remission, remission rate between 50 % and 75 %; level 3: obvious remission, remission rate over 75 %) with a lower level indicating less responsiveness to dopaminergic medication. According to their response to dopaminergic drugs the 79 patients were assigned to either the dopaminergic drug responsive fatigued (DDRF) group (fatigue amelioration rate > level 0) or the dopaminergic drug non-responsive fatigued (DDNRF) group (fatigue amelioration rate = level 0).

The demographic and clinical profiles were compared between the DDRF and DDRNF groups; the results of these comparisons are shown in Table 3. There were 34 (43.04 %) patients in the DDRF group and 45 (56.96 %) patients in the DDNRF group. Onset age was significantly higher in the DDNRF group than in the DDRF group, and significantly more severe depression was found in the DDNRF group. There were no significant differences between the two groups in age, sex, disease duration, education, disease severity (H-Y stage, UPDRS total score and all subscores), individual motor symptoms, sleep disturbance or cognitive functioning (Table 3).

As shown in Table 4, the Spearman rank order correlation coefficients revealed significant correlations between fatigue remission rate and age, onset age, and depression severity. Other factors including sex, education, LEDD, disease severity (H-Y stage, UPDRS total score and all subscales) and cognitive functioning were not associated with the fatigue remission rate.

To identify independent factors influencing dopaminergic responsive fatigue, variables significantly associated with fatigue remission rate in the bivariate correlation analyses (Table 4) were entered as independent variables into a logistic regression model. Only depression severity (odds ratio = 1.182, 95%CI 1.044–1.338, P = 0.008) was identified as an independent risk factor for dopaminergic drug non-responsive fatigue.

Sleep is necessary for the maintenance of mammalian homeostasis. Additionally, sleep is one of the most crucial determinants of fatigue severity [25]. In the general population, high-quality sleep is associated with less fatigue [26, 27]. Sleep disturbance has been associated with fatigue in several disorders, including cancer [28], multiple sclerosis [29] and traumatic brain injury [30]. In addition, it was reported that dissatisfaction with sleep, and not sleep itself, was associated with fatigue in psychotic patients [31].

Although the mechanisms of sleep abnormalities and fatigue have not yet been clarified, increased inflammatory cytokine release has been associated with poor sleep conditions and fatigue [32]. Specifically, the production of interleukin (IL)-1β and tumor necrosis factor (TNF)-α promotes non-rapid eye movement sleep (NREMS) under physiological and inflammatory conditions. Disturbed NREMS is associated with abnormal cyclic alternating pattern (CAP) rates and electroencephalogram arousals, and it is correlated with fatigue and sleepiness [33]. IL-1β and TNF-α alter neuronal excitability and induce symptoms associated with sleep loss by binding to their neural receptors, which subsequently regulate neurotransmitters or neuromodulators [34]. Moreover, the neurotransmitter serotonin governs sleep-wake behavior [35], and a reduction of serotonin transporters was found in subjects with fatigue syndrome [36]. These findings suggest that serotoninergic functions may play pivotal roles in controlling sleep and fatigue.

Fatigue, sleep disturbance and depression are the primary neuropsychiatric manifestations of PD, and they are related to patient quality of life [37, 38]. In this study, a logistic regression analysis revealed that sleep disturbance is the only independent factor for fatigue in PD patients, which is in accordance with the findings of a previous study [6]. Emerging lines of evidence indicate that sleep disorder treatment is effective for improving fatigue in patients with multiple sclerosis [39, 40]. Similarly, the management of sleep loss may provide a valuable approach for improving fatigue in PD patients. It has been proposed that fatigue stems at least in part from disturbed cytokine production in diseases such as PD [41]. Indeed, elevated secretion of IL-1β and TNF-α have been detected in the cerebrospinal fluid (CSF) of PD patients with sleep problems [42]. The circulating level of IL-1β is also increased in PD patients with fatigue [43]. Pavese et al. indicated that in addition to increased inflammatory cytokine production, fatigue in PD patients might result from serotonergic (5-hydroxytryptamine, 5-HT) dysfunction in the basal ganglia and limbic circuitry [44]. Considering the crucial roles that cytokine release and serotonergic functioning play in regulating sleep and fatigue, we hypothesize that sleep disturbance and fatigue may share a similar mechanism in the pathogenesis of PD, with abnormal cytokine release and serotonergic dysfunction being the primary components of this mechanism. Future studies should explore the molecular mechanisms involved in this process.

Another important finding of the current study was that dopaminergic medication improved fatigue in 43.04 % of PD patients with fatigue problems, whereas more than half of the fatigued patients were not responsive to dopaminergic medication. This result is consistent with ELLDOPA study [45], which showed that levodopa/carbidopa is effective in reducing the progression of fatigue in drug naïve PD patients [12]. Moreover, the study of ADAGIO [46] also showed that rasagiline was associated with significantly less progression of fatigue compared with placebo over a 9-month period in drug naïve patients with PD [47]. Both carbidopa/levodopa and rasagiline are dopaminergic drugs, which increase the extracellular levels of dopamine. Thus, we infer dopamine dysfunction may involve in the pathogenesis of fatigue in PD. The effects we observed might be due to enhanced dopaminergic neurotransmission, indicating the therapeutic value of dopaminergic treatment for fatigue relief in PD. However, according to the univariate logistic regression and correlation analysis performed in our study, there are no significant associations among fatigue, fatigue severity, remission rate, and dopaminergic medication dosage because a no dose-effect correlation was detected between levodopa and fatigue. Hence, levodopa may play a role in the mechanism underlying the development of fatigue, which is consistent with other current results. The dopamine imbalance hypothesis of fatigue was recently presented in multiple sclerosis and other neurological disorders [48]. Thus, restoring dopamine levels in the CNS by means of dopaminergic medication, such as levodopa, might be an essential strategy for the treatment of fatigue in PD.

Depression is another prominent non-motor symptom in PD [49]. Fatigue in PD is often associated with depression, using a logistic regression model, we further demonstrated that depression was the only independent factor for the efficacy of dopaminergic medication in treating fatigue in PD patients, which might imply crosstalk between depression, fatigue and dopamine insufficiency.

Pavese et al. found that compared with patients without fatigue, PD patients with fatigue showed greatly reduced serotonin transporter binding in the basal ganglia and limbic circuitry, as well as a significant reduction in dopamine uptake in the caudate and insula [44]. These results demonstrate that even though a dysfunctional serotonergic pathway plays a predominant role in fatigue, reduced dopaminergic functioning also contributes to the incidence of fatigue [44]. Hence, both dopaminergic and serotonergic pathways contribute to fatigue in PD patients. Interestingly, other studies have indicated that excessive 5-HT also induced fatigue. Indeed, increased 5-HT and decreased dopamine concentrations were detected in the brain in a rat model of exercise-induced fatigue [50]. These findings imply that the level of 5-HT may not be a determinant, and maintaining the balance between serotonergic and dopaminergic function is key for fatigue relief. Hence, dopamine administration may improve fatigue only in PD patients who have a functional serotonergic system with normal 5-HT concentrations and receptor functioning, which is consistent with our results. In other words, functional serotonergic pathway is necessary for dopamine to alleviate fatigue. Therefore, both 5-HT and dopamine should be considered during the treatment of PD fatigue. For instance, in the early stage of PD, a supplement of only 5-HT without levodopa medication may be insufficient for relieving fatigue, as levodopa is usually not prescribed early in PD due to the absence of obvious motor symptoms. Similarly, when fatigue is present with no obvious depression, a combined administration of levodopa with 5-HT, but not levodopa alone, may provide greater relief from fatigue. Thus, we hypothesize that dopaminergic medication is required, but not sufficient, for fatigue suppression in fatigued PD patients with moderate depression, and such treatment can restore serotonergic neurotransmission and serve as a combination therapy. This may offer an ideal strategy for the treatment of fatigue in PD patients.

Written informed consent was obtained from the patient for the publication of this report and any accompanying images.