|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Clinical Factors Associated With Cancer-Related Fatigue in Patients Being Treated for Leukemia and Non-Hodgkin’s LymphomaByFrom the Department of Symptom Research, Division of Anesthesiology and Critical Care, Hematology, Division of Medical Oncology, and Department of Biostatistics, The University of Texas M.D. Anderson Cancer Center, Houston, TX. Address reprint requests to Xin Shelley Wang, MD, Department of Symptom Research, Box 221, The University of Texas M.D. Anderson Cancer Center, 1100 Holcombe Blvd, Houston, TX 77030; email: swong@ mdanderson.org.
PURPOSE: To describe fatigue severity, fatigue interference, and associated factors in hematologic malignancies. PATIENTS AND METHODS: Patients being treated for leukemia and non-Hodgkin’s lymphoma (n = 228) completed the Brief Fatigue Inventory to rate fatigue severity and functional interference caused by fatigue. Data on patient demographics, Eastern Cooperative Oncology Group performance status, other physical symptoms, current treatments, and laboratory values were also collected. Descriptive statistics, bivariate correlation, and logistic regression were used for data analysis. RESULTS: Fifty percent of the sample reported severe fatigue, which was defined as a "fatigue worst" rating of 7 or greater. More patients with acute leukemia (61%) reported severe fatigue compared with those with chronic leukemia (47%) and non-Hodgkin’s lymphoma (46%). Increased fatigue severity significantly compromised patients’ general activity, work, enjoyment of life, mood, walking, and relationships with others. Fatigue severity was strongly associated with performance status, use of opioids, blood transfusions, gastrointestinal symptoms, and sleep disturbance items, as well as with low serum hemoglobin and albumin levels. Regression analysis indicated that nausea was the significant clinical predictor of severe fatigue (odds ratio, 13), and low serum albumin was the significant laboratory value predictor (odds ratio, 3.8). CONCLUSION: Disabling fatigue occurs with high frequency in hematologic malignancy, supporting a need to develop better methods of fatigue management. Better control of gastrointestinal and other symptoms may be of benefit. The mechanism and relationship between low albumin and severe fatigue needs to be investigated further, and longitudinal studies of the effects of treatment, host factors, and other symptoms are needed.
CANCER-RELATED fatigue has been reported to be highly prevalent in population-based surveys1 and in studies of some specific cancers and treatments.2-4 With the exception of recording fatigue as a toxicity measure, fatigue has not been assessed in a standard way, hindering comparison of its prevalence across diseases or treatments. Severe fatigue has been reported more frequently in patients with hematologic malignancies than in patients with solid tumors.5 Fatigue is the most common presenting symptom at diagnosis in those with chronic myelocytic leukemia (CML), occurring in at least one third of patients.6 Cancer-related fatigue is prominent in patients who have "B" symptoms of lymphoma, and it is listed in many textbooks as a common symptom in patients with systemic Hodgkin’s disease and non-Hodgkin’s lymphoma (NHL). Prolonged fatigue was reported by 29% of patients with CML treated with interferon.7 Long-term lymphoma survivors treated with autologous bone marrow transplantation had a high level of fatigue.4 Among those with advanced disease, more than 75% complained of diminished energy or fatigue.8 Fatigue was a predictor of poor survival in a study of patients with chronic lymphocytic leukemia.9 Fatigue can negatively affect patients’ compliance with treatment, can be dose limiting,10 and can significantly impair health-related quality of life.11,12 In the cancer population, fatigue is associated with psychological disturbance, symptom distress, and decreases in functional status.13 Anemia, both treatment related and disease related, has been shown to be a major contributor to increased fatigue and diminished quality of life in patients with cancer of many types.14 However, other potential mechanisms of debilitating fatigue have also been proposed, such as endocrine changes,15 physical deconditioning,16 and treatment-related and disease-related alterations of cytokines.17 At this point, little research has addressed the clinical factors that relate to fatigue or that predispose patients with hematologic malignancies to experience fatigue. When hematologists are confronted with questions about fatigue from cancer patients, they are often unable to provide information of how severe fatigue will become, how long it will last, what causes it, or what treatments might be effective for reducing it. Several advances have been made in the assessment of fatigue, allowing for the identification of patients who have fatigue so severe that it significantly impairs their function.5 By use one such validated fatigue assessment instrument, the Brief Fatigue Inventory (BFI), and a method of categorizing fatigue severity,18 we examined the prevalence, severity, and effect of fatigue in patients with acute (AL) and chronic leukemia (CL) and NHL. We also evaluated the association of severe fatigue with clinical and laboratory values expected to be related to fatigue. We used demographic information as well as the presence or absence of other physical symptoms, Eastern Cooperative Oncology Group (ECOG) performance status, and current treatments. On the basis of the fatigue literature, we expected that fatigue severity might be greater in older or female patients19 or in patients with poor performance status.18 In addition to fatigue, patients often report other symptoms caused by their disease or treatment, such as pain, infection, gastrointestinal (GI) symptoms, sleep disturbances, and body weight loss. We expected that those with a greater symptom burden would be more tired. A relationship between fatigue and low hemoglobin has already been reported.2,14 We were also interested in the relationship of other laboratory values reflecting disease progress or poor prognosis that could be associated with fatigue in leukemia and NHL, including low albumin, higher lactic dehydrogenase (LDH), measures of liver and renal dysfunction, and low blood counts caused by bone marrow suppression from chemotherapy. All of these variables could be potentially predictive of severe fatigue. Once these associations had been evaluated, we were interested in developing models that might help clinicians use this information to predict those patients most at risk for severe fatigue. We also hoped to identify variables that would be candidate variables for longitudinal studies of the relationship of fatigue with hematologic disease and treatment.
Sample This cross-sectional study was approved by the institutional review board of the University of Texas M.D. Anderson Cancer Center. Recruited patients were being treated in the Departments of Leukemia, Lymphoma, and Blood and Marrow Transplantation at this cancer center. A convenience sample of cancer patients (n = 246) was approached in both outpatient clinics and inpatient units. A research nurse explained the study to patients and assessed their eligibility. Patients were eligible if they were 18 years or older, had a pathologic diagnosis of leukemia or lymphoma, and were able to read and understand self-report questionnaires in English. Participating patients provided written informed consent.
Data Collection Patients rated several other symptoms on 0 to 10 scales, 10 being "as bad as you can imagine." These included "pain at its worst" during the past 24 hours, "ability to think clearly," "feeling rested and refreshed on awakening," "woke up during the night," "having difficulty falling asleep at night," and "woke up too early in the morning." Research nurses interviewed patients if they were having any of the following GI symptoms at the time of survey (yes/no): nausea, vomiting, loss of appetite, chewing or swallowing problems, and diarrhea. Patients were also asked whether their pain had been present for more than 3 months, their usual and current body weight, and whether or not they had evidence of infection during the past 2 weeks. Information about patients’ disease, current treatment, and laboratory values during the same week of the survey was recorded on the checklist. Disease data included diagnosis, ECOG performance status, disease status (ie, remission or relapse), and inpatient or outpatient status. Anticancer treatment data included whether the patient had received chemotherapy, radiotherapy, surgery, biotherapy, or bone morrow transplantation in the past month. Other intervention data included whether the patient had received supportive therapies (ie, blood transfusions or opioids for pain) during the past month.
Statistical Analysis
Data on fatigue severity, fatigue interference, and other variables were categorized and summarized by diagnosis: AL, CL, and NHL. Initial examinations of the relationships of fatigue and clinical variables (demographic, disease severity, other symptoms, and cancer treatments) and laboratory values were explored by using descriptive statistics, calculating bivariate correlations, and examining scatterplots. To compare the three diagnosis groups, we performed analysis of variance when the dependent variable was continuous and a Pearson’s We were interested in identifying predictors of severe fatigue. By use of recorded fatigue severity (severe v nonsevere) as the dependent variable, we developed two logistic regression models. The first model examined fatigue severity related to the patient’s clinical profile. The second model examined the relationship of fatigue with laboratory values.
The procedure to determine the predictor variables was conducted in two stages. First, variables were screened as candidate predictors for the regression model on the basis of the results of univariate analyses (two-way associations). A correlation between fatigue severity and clinical or laboratory variables had to be Second, stepwise regression was performed. Model fit was examined by plotting the residuals. In addition, the plots of the predicted values against the residuals were also investigated. Multicollinearity diagnostics were examined by calculating the collinearity statistic tolerance and the corresponding variance inflation factor. Tolerance should be close to unity, whereas the variance inflation factor should be relatively low. The results from the diagnostic test indicate that the degrees of multicollinearity among the independent variables were within acceptable limits. Models were also compared on the basis of the magnitude of the standardized coefficients and not solely on the P values because the groups had different sample sizes. All estimation procedures (forward conditional, likelihood ratio, and Wald’s procedure) resulted in the same set of predictors. Odd ratios and 95% confidence interval are reported.
Sample Characteristics We approached 246 patients, including 57 patients with a diagnosis of AL (35 with acute nonlymphoblastic leukemia and 22 with acute lymphoblastic leukemia), 49 patients with a diagnosis of CL (33 with CML and 16 with chronic lymphocytic leukemia), and 122 patients with a diagnosis of NHL. There were also 10 patients diagnosed with Hodgkin’s disease who were not included in this report because of their relatively small number. Eight patients refused to participate in the study, yielding a final sample of 228. Table 1 lists the demographic characteristics of the total sample and of the three major diagnostic groups.
Table 2 lists variables related to disease, treatment within the past month, and laboratory results during the week of the study in the total sample and for the three diagnostic groups. Sixty percent of patients in the study were inpatients. We categorized ECOG performance status scores into "good" (scores of 0 to 1) and "poor" (scores of 2 to 4). By this definition, 56% of the sample had a poor performance status at the time of the study. The mean ECOG performance status score for the total sample was 1.51 (SD, 0.96). Seventy-two percent of patients were currently not in remission, whereas 47% of patients had already relapsed at least once. Significantly more patients with CL were not in remission and not relapsed compared with patients with AL and NHL (all P < .05).
All patients in this study reported that they had received at least one type of cancer treatment during the previous month. The majority of patients (72%; n = 164) had received chemotherapy, and a few had received other cancer treatments during the previous month (Table 2). Significantly more patients with AL than NHL had had chemotherapy (P = .017). Forty-seven percent of patients had received biotherapy within the past month (eg, granulocyte colony-stimulating factor or interferon). Patients had also received other interventions, including blood transfusions (55%; n = 125) and opioids for pain management (19%; n = 44) within the previous month. Six patients had had more than one transfusion, with one CL patient having had two and one AL patient having had three; three NHL patients had had two and one NHL patient had had four transfusions. Fewer patients with NHL than AL (P = .000) or CL (P = .000) had had blood transfusions.
Fatigue Prevalence and Severity
Effect of Severe Fatigue on Function
Comparing the mean scores of interference items across disease groups, patients with AL reported significantly greater impairment of their fatigue-related mood, thinking ability, and relationships with others than other groups (all P < .05). AL patients also had a significant higher level of impaired walking ability, normal work, and enjoyment of life than patients with NHL (all P < .05). There was no significant difference for the interference with general activity item across groups.
Fatigue and Demographic Variables
Fatigue and Disease Severity Inpatient status was also highly associated with "fatigue worst" (r = .421; P < .001). Inpatients reported significantly greater levels of fatigue than outpatients (7.1 v 4.8; P < .001). Compared with outpatients, a greater percentage of inpatients had poor ECOG performance status (84% v 15%; P < .001), had received chemotherapy (84% v 53%; P < .001), had received blood transfusions in the past month (72% v 33%; P < .001), had severe pain (59% v 42%; P < .05), and were receiving opioids (27% v 9%; P < .01). Inpatients also had significantly lower levels of hemoglobin (9.8 v 11.0 g/dL; P < .001) and albumin (3.1 v 3.8 g/dL; P < .001) than outpatients. Patients in remission (28%) reported less severe fatigue than patients who were not in remission (5.4 v 6.4; P < .05). Similarly, patients who had relapsed reported significantly greater levels of fatigue than patients who had not relapsed (6.6 v 5.7; P < .05).
Fatigue and Its Relationship With Other Symptoms
It is interesting to note that fatigue severity was not correlated with the number of hours that patients had slept during the past 24 hours (r = .162). Forty-eight percent of patients reported sleeping from 6 to 10 hours during the past 24 hours (17% reported 2 to 5 hours, and 19% reported 11 to 21 hours; mean, 8.4 hours, SD, 3.66 hours).
Fatigue and Its Relationship With Cancer Treatment
Fatigue and Laboratory Variables Hemoglobin. In this sample, 87% of female patients and 94% of male patients had anemia by the diagnostic criteria of anemia at the M.D. Anderson Cancer Center (hemoglobin concentrations of < 12 g/dL in women or < 14 g/dL in men). The mean level of hemoglobin in the total sample was 10.3 g/dL (7.1 to 15.9 g/dL; SD, 1.7). Hemoglobin level was significantly different across diagnoses (P = .000), with NHL patients having significantly higher hemoglobin levels than patients with AL (P = .000). As expected, hemoglobin level was negatively correlated with "fatigue worst" (r = -.332; P < .001). The mean level of hemoglobin was significantly lower in patients with severe fatigue than nonsevere fatigue (9.9 v 10.8 g/dL; P < .001). Fifty-four percent of patients with severe fatigue had hemoglobin levels of less than 10 g/dL. Albumin. Forty-eight percent of patients had serum albumin levels below the reference range, on the basis of the criteria used at the M.D. Anderson Cancer Center (reference range, 3.5 to 4.7 g/dL). The mean level of albumin in the total sample was 3.4 (1.7 to 4.7; SD, 0.6). Albumin levels were significantly different across diagnoses (P = .002), with NHL patients having significantly higher levels of albumin than patients with AL (P = .004). Hemoglobin was the only laboratory value significantly correlated with albumin (r = .435; P < .001). Patients who had had chemotherapy had lower mean albumin (P < .01). Fatigue severity was negatively correlated with albumin level (r = -.396; P < .001). Patients with albumin levels lower than the reference range reported significantly higher levels of fatigue than patients with albumin levels within the reference range ("fatigue worst," 7.1 v 5.3; P < .001). Figure 1 presents the percentage of patients with severe fatigue by albumin level.
Other laboratory variables. LDH positively correlated with fatigue in the NHL group (r = .347; P < .001), but not in the other groups (r = .137 in the AL group; r = .058 in the CL group). The mean fatigue score was significantly higher in patients with a higher than normal range (618 IU or higher) of LDH (7.0 v 5.5; P < .001). There was no observed correlation of fatigue and WBC in any disease group. Platelet count was correlated with fatigue in the CL group (r = -.309; P < .05), but not in the other groups.
Predictors of Severe Fatigue by Diagnosis
Patients with AL. Patients with AL were more likely to report severe fatigue if they were disabled or unemployed (r = .322; P < .05) or inpatients (r = .264; P < .05) or if they reported any of the following symptoms: any GI symptom (r = .319 to .650; all P < .05), woke up too early in the morning (r = .287; P < .05), woke up during the night (r = .278; P < .05), or reported difficulty in thinking clearly (r = .430; P < .05). These variables were included as candidate predictors in logistic regression model 1 to predict severe fatigue in patients with AL. In this model, nausea was the predictor of severe fatigue. Patients with nausea had a significantly higher level of fatigue than patients without nausea (7.7 v 4.8; P < .001). Low serum albumin was the laboratory value correlated with fatigue (r = .245; P < .05), and it was the laboratory value that predicted severe fatigue in model 2 in AL patients. Patients with CL. Patients with CL were more likely to report severe fatigue if they had poor ECOG performance status (r = .341; P < .05), had received a blood transfusion within the past month (r = .292; P < .05), or reported any of the following symptoms: any GI symptom (r = .215 to .476; P < .05 for nausea only), insomnia (r = .330; P < .05), or difficulty in thinking clearly (r = .428; P < .05). Of all previously mentioned candidate predictors in logistic regression model 1 to predict severe fatigue in patients with CL, the significant predictor was nausea. Patients with nausea in this group had significantly higher fatigue levels compared with patients without nausea (7.2 v 4.5; P < .005). Of all the laboratory values, only albumin and platelets were significantly correlated with "fatigue worst" (r = -.298; r = -.309; both P < .05). However, there were no significant predictors of severe fatigue in model 2 for this group. Patients with NHL. Patients with NHL were more likely to report severe fatigue if they had poor ECOG performance status (r = .491; P < .001), were inpatients (r = .414; P < .001), had received blood transfusions during the past month (r = .255; P < .05), were currently receiving opioids (r = .336; P < .001), or reported any of the following symptoms: any GI symptom (r = .363 to .526; P < .001), not feeling rested and refreshed on awakening (r = .262; P < .05), waking up during the night (r = .236; P < .05), insomnia (r = .265; P < .05), moderate to severe pain (r = .236; P < .05), decrease in body weight (r = .291; P < .001), or difficulty in thinking clearly (r = .574; P < .001). The significant predictor from the previously mentioned clinical candidate predictors in logistic regression model 1 in this group was also nausea. Patients with nausea in this group had significantly higher fatigue levels compared with those without nausea (7.3 v 4.3; P < .001). The laboratory values albumin, hemoglobin, and LDH were significantly correlated with fatigue (r = -.478, -.335, and .347, respectively; all P < .001) in patients with NHL. These variables were the candidate predictors in regression model 2 to predict severe fatigue in NHL patients. The significant predictor was albumin. Combined model of severe fatigue. Variables that were used for this combined model (model 3) were nausea, albumin, and an interaction term involving nausea and albumin. This was performed to adjust simultaneously for the effect of both clinical and laboratory predictors and to test whether the likelihood of severe fatigue was the same for different levels of albumin for patients with and without nausea. Because there was no predictor for the CL group with the laboratory model, we performed the logistic regression only on the AL and the NHL groups. For the AL group, only nausea was a significant predictor of severe fatigue. A patient with nausea was 24 times more likely to have severe fatigue. For the NHL group, both nausea and albumin levels were significant predictors of severe fatigue. A patient with low albumin was four times more likely to have severe fatigue. In addition, a patient with nausea was five times more likely to have severe fatigue. The interaction between albumin and nausea was not statistically significant for the AL and NHL groups.
By using a previously well validated fatigue measure, the BFI, this study demonstrated that approximately half of patients receiving treatment for leukemia and NHL report severe fatigue that substantially interferes with normal function. In another study, we found that less than 10% of a community-dwelling sample, including persons with cancer and other chronic diseases, had fatigue that could be classified as severe.18 Previous studies performed with patients who had solid tumors have identified fatigue as a significant symptom during the course of treatment.1,22,23 As was found here, these previous studies found that fatigue had a substantially negative effect on quality of life in general.1,24 One component of the National Cancer Center Network definition of cancer-related fatigue is its unusual severity.25 A second component is that cancer-related fatigue interferes with usual function. In this study, those with severe fatigue reported substantially more fatigue-related interference than those categorized as having nonsevere fatigue. Across all three diagnostic groups, the majority of the mean interference scores for the severe-fatigue group were higher than those of the nonsevere group by more than 1 SD of the mean of the nonsevere group. The severely fatigued group also reported significantly greater interference with thinking clearly. Thus, this study documents the high risk that patients being treated for hematologic malignancies have for disabling fatigue. Among the diagnostic groups in this study, the AL group had a higher percentage of patients with severe fatigue and, as would be expected, reported a greater level of fatigue and fatigue-related functional impairment than the NHL group. Although the performance status of the AL group was not different from that of other groups, more AL patients had recently received chemotherapy than NHL patients. Patients with AL also had lower levels of hemoglobin and albumin than NHL patients. There are several limitations to this study. First, information on the nonfatigue symptoms that patients experienced was obtained by asking whether these symptoms were present or absent. Asking for ratings of the severity of these symptoms in the same 0 to 10 format used for pain and fatigue would have helped to clarify the relationship between the severity of these symptoms and the severity of fatigue. Second, there was no attempt to determine how the affective status of patients related to how tired they were. An association between fatigue and depression in patients with cancer has been found in some studies, although one study found that the expression of fatigue and depression, if they are present, follows a different time course, suggesting that depression may not play a causal role in cancer-related fatigue.26 Finally, this study is limited by its cross-sectional design. It is reasonable to expect that fatigue will vary over the course of treatments. Having patients complete measures similar to the ones used in this study before, at several time points during, and after cancer therapy should give us information about the onset and persistence of severe fatigue. Also, a longitudinal design would provide a better way of describing the relationship of fatigue, nausea, and low albumin. Several variables that have been found to be associated with fatigue in noncancer populations, such as age and sex, were not found to be related to severe fatigue in this sample. Other demographic variables, including education, marital status, and ethnic heritage were also not predictive of severe fatigue, although the range of most of these variables was restricted. As expected, patients with less adequate performance status were at greater risk for severe fatigue. Also, remission was associated with a lesser, and relapse was associated with a greater, probability of severe fatigue. These results, coupled with previous research indicating the high prevalence of fatigue in patients with advanced cancer and the prognostic value of fatigue in predicting survival, suggest the prominent role of disease progression in the development of cancer-related fatigue. In this study sample, patients currently receiving chemotherapy (72%) reported similar levels of fatigue compared with patients not currently receiving chemotherapy, suggesting that patients may experience fatigue caused by both disease and treatment. Patients reported significantly higher levels of fatigue when they experienced other symptoms related to disease or treatment, such as pain, nausea and other GI symptoms, and sleep disturbance. It is interesting that duration of sleep was not correlated with fatigue, whereas sleeping pattern changes were associated with fatigue. There was also an association of severe fatigue with the use of opioid medication for pain control, because of the fatiguing effects of pain itself, the sedating effects of these medications, or both. Approximately one third of the sample had evidence of recent infection or greater than 10% of weight loss. These variables were not correlated with fatigue. In the multivariate models, nausea was the clinical variable that predicted severe fatigue in patients with leukemia and NHL (classification rate at 74% to 84%). Vomiting was not a predictor. Approximately 60% of patients usually develop nausea, and 50% reported vomiting after chemotherapy.27 Patients with NHL undergoing cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy judged nausea to be the "most troublesome" problem, followed by fatigue, taste change, constipation, and difficulty sleeping.28 Current available pharmacologic agents are unable to provide complete protection from either anticipatory or posttreatment nausea and vomiting, and these symptoms have remained prevalent despite the use of increasingly potent antiemetic medications.29 There are many causes of nausea that could be identified and corrected, such as metabolic disturbances, pain, and medications, including antibiotics, bronchodilators, and opioids.30 It is more difficult to control nausea than vomiting clinically. However, nausea is often best treated by a multidisciplinary approach, including behavioral intervention and patient education.29 It would be interesting to examine changes in fatigue in the context of a trial in which nausea was being aggressively treated. It would also be interesting to study the interaction of fatigue and nausea during treatment by using the longitudinal design. Low albumin was the laboratory value that predicted severe fatigue in a regression model examining laboratory variables, although hemoglobin was also highly correlated with severe fatigue. The association between albumin and fatigue was strong in all three disease groups. The mechanism or basis for this association needs further research. Of course, low albumin could indicate poor nutritional status caused by progressive disease, GI toxicity from chemotherapy, or both. It might also be partially explained as being related to changes in cytokines as an acute-phase response to the disease and to aggressive treatment. Albumin is the principal negative acute-phase reactant in humans, and its reduction can reflect an ongoing acute-phase protein response.31 It has been demonstrated that proinflammatory cytokines such as tumor necrosis factor and interleukin (IL)-6 are known to downregulate albumin synthesis and increase acute-phase protein production.31 There is also some evidence of cytokine-mediated phenotypic features in leukemia and lymphoma, including hypoalbuminemia. For example, IL-6 is a potent immune modulatory cytokine that may have pathogenetic significance in several malignancies. High IL-6 levels have been associated with a poor prognosis in NHL, advanced Hodgkin’s lymphoma, and other cancers.32 In patients with aggressive NHLs, increased serum levels of IL-6 at diagnosis are frequent, and serum IL-6 levels are strongly associated with adverse disease features, such as B symptoms, poor performance status, and low serum albumin level (r = -.62; P < .0001).33 Although a relationship between fatigue and cytokines has not been established, cytokine changes may play a role in the pathophysiology of cancer-related fatigue. Further research on the relationships among albumin, fatigue, and cytokines (especially proinflammatory cytokines) might produce leads in understanding why patients become severely fatigued. There was little surprise in finding a strong relationship between anemia and cancer-related fatigue. Treatment- or disease-related anemia is often associated with fatigue in cancer patients, and reducing anemia has been shown to reduce fatigue.14 The etiology of anemia in cancer patients is multifactorial. It may be the result of chronic disease, myelosuppressive chemotherapy or radiation therapy, low levels of erythropoietin, and less marrow responsiveness to the available erythropoietin, blood loss, or direct bone marrow infiltration by the tumor.34,35 In this study setting, the criterion for transfusion was a hemoglobin level of 7 g/dL, which would explain the high correlation of transfusions with fatigue. The reason that anemia was not a predictor of severe fatigue in the multivariate model is most likely that most patients were anemic when studied, constricting the range of this variable. There was no observed relationship of fatigue severity and total WBC count, but either leukopenia or pathologic leukocytosis might be related to increased fatigue. LDH level is a conventional prognostic factor that reflects tumor burden in lymphoma. Its correlation with fatigue is more evidence of the effect of disease progression on fatigue severity. Thrombocytopenia is often found in patients with CL, and its correlation with fatigue could be explained by side effects of treatment. The development of methods for measuring cancer-related fatigue gives us the tools necessary to conduct the descriptive and epidemiologic studies that define who will have the most severe fatigue and what correlates with the development of fatigue. These methods also allow us to recognize the significant burden that fatigue places on those with hematologic and other malignancies and point the way to understanding the underlying mechanisms.
Supported by American Cancer Society Special Institute Grant grant no. 21, grant no. CA 26582 awarded by the National Cancer Institute, Public Health Service, Bethesda, MD, and an unrestricted educational grant from Ortho-Biotech, Raritan, NJ.
1. Vogelzang NJ, Breitbart W, Cella D, et al: Patient, caregiver, and oncologist perceptions of cancer-related fatigue: Results of a tripart assessment survey—The Fatigue Coalition. Semin Hematol 34: 4-12, 1997[Medline]
2.
Portenoy RK, Itri LM: Cancer-related fatigue: Guidelines for evaluation and management. Oncologist 4: 1-10, 1999 3. Smets EM, Visser MR, Willems-Groot AF, et al: Fatigue and radiotherapy: (A) Experience in patients undergoing treatment. Br J Cancer 78: 899-906, 1998[Medline] 4. Knobel H, Loge JH, Nordoy T, et al: High level of fatigue in lymphoma patients treated with high dose therapy. J Pain Symptom Manage 19: 446-456, 2000[CrossRef][Medline] 5. Cleeland CS, Wang XS: Measuring and understanding fatigue. Oncology 13: 91-97, 1999[Medline] 6. Savage DG, Szydlo RM, Goldman JM: Clinical features at diagnosis in 430 patients with chronic myeloid leukemia seen at a referral centre over a 16-year period. Br J Haematol 96: 111-116, 1997[CrossRef][Medline] 7. Cortes J, Kantarian HO, Robertson LE, et al: Result of interferon-alpha therapy in patients with chronic myelogenous leukemia 60 years of age and older. Am J Med 100: 452-455, 1996[CrossRef][Medline] 8. Curtis EB, Kretch R, Walsh TD: Common symptoms in patients with advanced cancer. J Palliat Care 7: 25-29, 1991[Medline] 9. Chen PM, Lin SH, Fan SF, et al: Genotypic characterization and multivariate survival analysis of chronic lymphocytic leukemia in Taiwan. Acta Haematol 97: 196-204, 1997[Medline] 10. Smith MJ, Khayat D: Psychological side effects induced by interleukin-2/alpha interferon: Clinical observations, measurement, biology, treatment. 4th International Congress on Anti-Cancer Chemotherapy Meeting, Paris, France, February 2-5, 1993 (abstr) 11. Schwartz JE, Jandorf L, Krupp LB: The measurement of fatigue: A new instrument. J Psychosom Res 37: 753-762, 1993[CrossRef][Medline] 12. Portenoy RK, Miaskowski C: Assessment and management of cancer-related fatigue, in Berger A, Portenoy RK, Weissman DE (eds): Principles and Practice of Supportive Oncology. Philadelphia, PA, Lippincott, 1998, pp 109-118 13. Glaus A: Assessment of fatigue in cancer and non-cancer patients and in healthy individuals. Support Care Cancer 1: 305-315, 1993[CrossRef][Medline]
14.
Glaspy J, Bukowski R, Steinberg D, et al: Impact of therapy with epoetin alfa on clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in community oncology practice: Procrit Study Group. J Clin Oncol 15: 1218-1234, 1997 15. Jones TH, Wadler S, Hupart KH: Endocrine-mediated mechanisms of fatigue during treatment with interferon-alpha. Semin Oncol 25: 54-63, 1998 16. Dimeo FC, Stieglitz RD, Novelli-Fischer U, et al: Effects of physical activity on the fatigue and psychologic status of cancer patients during chemotherapy. Cancer 85: 2273-2277, 1999[CrossRef][Medline] 17. Stone P, Richards M, Hardy J: Fatigue in patients with cancer. Eur J Cancer 34: 1670-1676, 1998 18. Mendoza TR, Wang XS, Cleeland CS, et al: The rapid assessment of fatigue severity in cancer patients: Use of the Brief Fatigue Inventory. Cancer 85: 1186-1196, 1999[CrossRef][Medline] 19. Cannon JC, Pierre BA: Gender differences in host defense mechanisms. J Psychiatr Res 31: 99-113, 1997[CrossRef][Medline] 20. Hosmer DW, Lemeshow S: Applied Logistic Regression. New York, NY, Wiley, 1989 21. Serlin RC, Mendoza TR, Nakamura Y, et al: When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain 61: 277-284, 1995[CrossRef][Medline] 22. Irvine DM, Vincent L, Bubela N, et al: A critical appraisal of the research literature investigating fatigue in the individual with cancer. Cancer Nurs 14: 188-199, 1991[Medline] 23. Portenoy RK, Thaler HT, Kornblith AB, et al: Symptom prevalence, characteristics and distress in a cancer population. Qual Life Res 3: 183-189, 1994[CrossRef][Medline]
24.
Kroenke K, Wood DR, Mangelsdorff AD, et al: Chronic fatigue in primary care: Prevalence, patient characteristics, and outcome. JAMA 260: 929-934, 1988 25. Atkineson A, Barbservick A, Cella D, et al: NCCN practice guidelines for cancer-related fatigue. Oncology 14: 151-161, 2000[Medline] 26. Visser MRM, Smets EMA: Fatigue, depression and quality of life in cancer patients: How are they related? Support Care Cancer 6: 101-108, 1998[CrossRef][Medline] 27. Morrow GR, Roscoe JA, Hickok JT: Nausea and Vomiting: Management of Specific Symptoms—Psycho-Oncology. New York, NY, Oxford University Press, 1998, pp 476-484 28. Sitzia J, North C, Stanley J, et al: Side effects of CHOP in the treatment of non-Hodgkin’s lymphoma. Cancer Nurs 20: 430-439, 1997[Medline] 29. Morrow GR: Behavioural factors influencing the development and expression of chemotherapy induced side effects. Br J Cancer Suppl 19: S54-S60, 1992[Medline] 30. Skeel RT, Tipton J: Special issues: Symptom management, in Brain MC, Carbone PP (eds): Current Therapy in Hematology-Oncology. St Louis, MO, Mosby, 1995, pp 578-587 31. Fearon KC, Falconer JS, Slater C, et al: Albumin synthesis rates are not decreased in hypoalbuminemic cachectic cancer patients with an ongoing acute-phase protein response. Ann Surg 227: 249-254, 1998[CrossRef][Medline] 32. Seymour JF, Talpaz M, Hagemeister FB, et al: Clinical correlates of elevated serum levels of interleukin 6 in patients with untreated Hodgkin’s disease. Am J Med 102: 21-28, 1997[Medline]
33.
Seymour JF, Talpaz M, Cabanillas F, et al: Serum interleukin-6 levels correlate with prognosis in diffuse large-cell lymphoma. J Clin Oncol 13: 575-582, 1995 34. Miller CB, Jones RJ, Piantadosi S, et al: Decreased erythropoietin response in patients with the anemia of cancer. N Engl J Med 322: 1689-1692, 1990[Abstract] 35. Spivak JL: Cancer-related anemia: Its causes and characteristics. Semin Oncol 21: 3-8, 1994[Medline] Submitted August 25, 2000; accepted November 2, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
6) fatigue on the BFI "fatigue worst" item on the basis of our previous study.
2 test when the dependent variable was categoric. The omnibus analysis of variance test was followed up by using Tukey’s test. A significant 
0.25 with the "fatigue worst" rating (a clinically important relationship that we determined a priori). Because of the differences in sample size, we decided to examine effect sizes rather than the usual criterion of using the P value. If a variable had a marginal association based on a P value
