Annals of Oncology 8: 1251^1255, 1997. ß1997Kluwer Academic Publishers. Printed in the Netherlands. Original article ______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Correlation between physical performance and fatigue in cancer patients F. Dimeo,1 R.-D. Stieglitz,2 U. Novelli-Fischer,2 S. Fetscher,3 R. Mertelsmann3 &J.Keul1 1Department of Rehabilitation, Prevention and Sports Medicine, 2Department of Psychiatry and Psychotherapy, 3Department of Hematology/ Oncology, Freiburg University Medical Center, Freiburg, Germany Summary physical performance was found (r =º0.30;P 5 0.01). How- ever, intensity of fatigue showed a strong correlation with Background: Fatigue and a reduction in physical ability are several indicators of psychological distress such as depression common and often severe problems of cancer patients regard- (r = 0.68), somatization (r = 0.64) and anxiety (r =0.63;P for less of disease stage and modality of treatment. However, while all 50.001). Furthermore, patients with lower levels of phys- physical performance can be assessed objectively with labora- ical performance had signi¢cantly higher scores for depression tory tests, fatigue is a subjective phenomenon whose percep- (P = 0.005), somatization (P = 0.03) and anxiety (P = 0.08), tion is in£uenced by past experience and expectations for the and signi¢cantly lower scores for vigor (P = 0.05) than their future. counterparts whose physical performance was higher. Patients and methods: To evaluate the correlation between Conclusions: We conclude that fatigue in cancer patients fatigue and physical impairment, we assessed maximal phys- may be related to mood disturbance but appears to be inde- ical performance with a treadmill test, and mental state with pendent of physical performance. Moreover, low physical per- two questionnaires, the Pro¢le of Mood States (POMS) and formance can be viewed as an independent predictor of mental the Symptom Check List (SCL-90-R), in a successive series of distress in cancer patients. 78 cancer patients with solid tumors or hematological malig- nancies. Key words: cancer fatigue, physical performance, psychological Results: A weak association between fatigue and maximal distress Introduction maximal workload in km/h or watts). Fatigue, on the other hand, has to be assessed indirectly by self-ratings. Fatigue ¢gures among the most common problems of Another factor complicating evaluation of fatigue in cancer patients. According to several studies, this symp- cancer patients is that ^ as with every subjective phe- tom a¡ects up to 70% of patients during chemo- and nomenon ^ perception of fatigue strongly depends on radiotherapy [1^3]. It has been reported that impair- past experience. Since cancer-related decreases in phys- ment of physical functioning may persist for years after ical activity can persist for a long time, perception of cessation of therapy in up to 30% of cancer survivors [3^ fatigue can change as it becomes chronic. It has been 5]. For many cancer patients, fatigue is severe and im- suggested that cancer patients gradually become accus- poses limitations on normal daily activities. Postulated tomed to their impaired physical condition and ¢nally etiological mechanisms for the development of fatigue experience it as normal [7]. Furthermore, the subjective include anemia, impaired nutritional status, sleep dis- and objective evaluation of the limitation in physical turbances, biochemical changes secondary to disease performance can produce di¡erent results: a substantial and treatment, psychosocial factors, and a reduced level discordance has been observed in estimations of func- of activity [6]. However, the causes of the impairment of tional ability with the Karnofsky index by patients and physical functioning in this setting are not yet fully medical sta¡ [8]. Nevertheless, in some clinical situa- understood. tions fatigue can exist independently of deterioration in In fact, fatigue represents only one aspect of the physical performance; for example, fatigue is frequently problem of physical impairment experienced by cancer a symptom in patients with depressive disorders whose patients. Cancer is usually accompanied by an `asthenic physical performance is not necessarily impaired. syndrome' consisting of two components, one objective In light of these considerations, we investigated the (loss of physical performance) and one subjective (fa- relationship between fatigue and physical performance tigue). This di¡erence considerably complicates the study in cancer patients. of asthenia in cancer patients. Physical performance can be directly determinated by laboratory tests and ex- pressed in units (maximal oxygen uptake in ml/kg/min, 1252 Table 1. Baseline data of patients in the study (n = 78). mental symptomatology in di¡erent patient populations and have shown high reliability and validity. Gender 48 women, 30 men Maximal oxygen consumption in ml O2 per kg of body weight per Age 40 þ 11.3 (18^60) minute was calculated according to the guidelines of the American Body mass index 23.9 þ 3.7 (18^34) College of Sports Medicine [13]. Number Percentage of patients Statistical analysis Disease To evaluate the association between intensity of fatigue and physical Breast cancer 25 32% performance, a partial correlation test, including age and gender as Metastatic breast cancer 9 11% independent covariables, was carried out. Employing the Spearman Testicular cancer 11 14% rank correlation test, in a further analysis we calculated the correlation Sarcoma 3 4% between fatigue and the remaining sub-scales of the POMS (vigor, Small-cell lung carcinoma 1 1% anger, and depression), and between fatigue and the sub-scales of the Multiple myeloma 2 2% SCL-90-R considered to be of relevance (somatization, depression, Hodgkin's disease 6 8% anxiety, hostility, and interpersonal sensitivity). Non-Hodgkin's disease 21 27% Maximal physical performance of the patients was compared with the values for aerobic power of healthy persons, obtained from stand- Maximal physical performance ard tables [13]. In these tables, maximal physical performance can be Very poor (50% of maximum) 49 62% assigned to one of six functional categories representing percentages of Poor (50%^54% of maximum) 6 8% the maximal oxygen uptake (VO2max) values recorded in healthy Fair (55%^65% of maximum) 7 9% persons. The six categories are: very poor (VO2max lower than 50% Good (66%^69% of maximum) 6 8% of the maximal observed value), poor (50%^54%), fair (55%^65%), Excellent (70%^75% of maximum) 8 10% good (66%^70%), excellent (71%^75%), and superior (higher than Superior (475% of maximum) 2 3% 76%). To evaluate the in£uence of physical performance on mental distress, the psychological scores of patients whose physical perfor- Maximal physical performance of the patients was classi¢ed in accord- mance was in the lowest categories (`very poor' and `poor') were ance with tables of normal values for aerobic power tests for healthy compared with the remaining patients by the Mann^Whitney ranks adults [13].Values are shown as mean þ standard deviation; ranges are test. To identify signi¢cant predictors of fatigue, a stepwise multiple shown in brackets. regression analysis was carried out. It included all POMS subscales, the SCL-90-R subscales considered relevant (depression, somatiza- tion, and anxiety) and maximal physical performance as independent Patients and methods variables, and fatigue as a dependent variable. All statistical calcula- tions were carried out using the Statistical Package for the Social A series of 89 successive cancer patients with solid tumors or hemato- Sciences (SPSS 6.1.2). A value of P 5 0.05 was considered to show logical malignancies were enrolled in the study (Table 1). Inclusion statistical signi¢cance, and a value of r 4 0.50 to indicate relevant criteria were: age between 18 and 60 years, active malignancy, an correlation. absence of associated psychiatric, muscular, cardiovascular or pulmo- nary disease, and the ability to understand written German. All pa- tients were considered for a high-dose chemotherapy (HDC) with peripheral autologous blood stem cell transplantation. In the weeks Results preceding hospital admission for HDC, all of them received one to 2 four chemotherapy cycles consisting of etoposide 500 mg/m , ifosfa- According to the guidelines of the American College of mide 4 g/m2, and cisplatinum 50 mg/m2, with or without epirubicin 50 mg/m2 (VIP/VIP-E). Sports Medicine, the physical performance of 55 of 78 After giving their informed consent, all patients underwent a tread- patients in the study (70%) was classi¢ed as `poor' or mill stress test for assessment of maximal physical performance. The `very poor' (Table 1). These patients had signi¢cantly test began with 3 km/h and 1.5% elevation; the speed was increased by more somatic complaints (P =0.03),weremorede- 1 km/h every third minute, while the elevation remained unchanged.