Fatigue assessment in Parkinson's disease
KATSAROU Ζ., BOSTANTJOPOULOU S., PEITSIDOU E., KOURTESI G., TSIPROPOULOU V., KAFANTARI A., KARACHRISTIANOU S.

Background: Fatigue is a major non-specific symptom in a number of disorders including Parkinson's disease (PD). Fatigue is a vague concept and although it is understood by all, it is hard to define and much more difficult to measure precisely. Usually, the term fatigue encompasses a number of concepts, including decreased mental and physical endurance, low motivation and performance that is inferior to what one would normally expect.

Objective: In this study we attempted to assess fatigue in Greek PD patients by means of a reliable English questionnaire: he Fatigue Severity Scale (FSS). Our first goal was to explore the reliability and validity of the Greek translation of this instrument and then to examine the relationship of fatigue to various clinical characteristics of PD.

Method: Fatigue was assessed in 100 patients with idiopathic Parkinson's disease (57 male, 43 female) and 127 matched for age and education normal controls (59 male, 68 female). PD patients were selected from the population of our Outpatient clinic for Movement Disorders, after excluding subjects who were cognitively impaired (Mini Mental State Examination score below 24), or had severe motor fluctuations. Other exclusion criteria were systemic diseases such as heart failure, thyroid dysfunction, autoimmune disorders, anemia, diabetes, kidney, pulmonary and hepatic insufficiency as well as cancer. Controls were recruited from the population of PD patients' relatives, visitors to the Hospital and attendees of regional Senior Centers. They were not formally tested for cognitive impairment, but they were questioned for the presence of memory and reasoning problems as well as for depression.

The English version of the FSS questionnaire was translated into Greek according to standard procedures and applied to all subjects. This brief one-dimensional questionnaire consists of nine statements. Each statement is rated by the examinee on a scale from 1 to 7 with 1 indicating "Strongly disagree" (e.g. absence of fatigue-related symptoms) and 7 indicating "Strongly agree". 4 is the neutral point. This score is the highest possible to exclude the presence of clinically relevant fatigue. All subjects filled out the questionnaire in the presence of an examiner who was instructed to offer an explanation if requested, but was not allowed to cue the subject with an answer.

The PD patients were screened for depression by means of the Beck Depression Inventory (BDI). They also filled out the MOS 36-Item Short-Form Health Survey (SF-36). Motor disability in PD patients was clinically evaluated by means of the motor score of the Unified Parkinson's Disease Rating Scale. PD patients were also classified according to the Hoehn and Yahr classification scale.

Results: According to Nunally's criterion, reliability analysis of all scale items yielded an excellent Cronbach's alpha for PD patients (alpha=0.90) and controls (alpha=0.84) . Item to total correlation coefficients were satisfactory for all items, ranging from 0.432 to 0.794. Correlation between the FSS score and SF-36 Energy/Fatigue subscale yielded a significant negative Pearson's product-moment correlation coefficient (r=-0.375, p=0.004), showing good concurrent validity ofthe Greek version of the FSS.

The PD patients' FSS score was higher than controls (4.83±1.63 versus 3.45±1.55, p=0.0001). We calculated a ROC curve for the FSS as well. The area under the curve (AUC) was moderately satisfactory.

(AUC=0.692; 95% confidence interval). Multivariate analysis between FSS score and various clinical parameters of the disease showed a significant relationship between fatigue severity and depression (p=0.009). However, after exploring the causative relationship of depression to fatigue further, we performed a separate comparison of FSS scores between a subgroup of patients without depression (BDI scores less than 13) and controls. Non-depressed PD patients still had significantly higher scores in FSS (mean score=4.7±1.16, p=0.001) compared to controls.

This elevated fatigue score in a non-depressed subgroup of PD patients failed to prove a direct causative relationship between fatigue and depression. In this subgroup high FSS scores were associated with motor disability.

Conclusions: Our study has shown that the Greek translation of the FSS is a reliable and valid instrument for assessing fatigue in Parkinson's disease. PD patients yielded significantly higher FSS scores than controls. After excluding depression as a confounding factor, we found that fatigue was associated with motor disability.

Key words: Fatigue, Parkinson's disease, depression, FSS scale.