Cancer-Related Fatigue On Daily Living Activities in Pediatric Cancer Survivors

Aim Cancer-related long term complications such as cardiovascular disease, fatigue, weight-related problems, and emotional disturbances are found to be increased in pediatric cancer survivors (PCS). The relationship between daily living activities (DLA) and such complications is still being investigated. Our aim in this study was to investigate the impact of cancer-related fatigue on DLA in PCS. Methods Cancer-related fatigue was assessed by “PedsQL Multidimensional Fatigue Scale (MFS)”, “Visuel Analogue Scale (VAS)” and “Borg Fatigue Scale”. The DLA was evaluated by WeeFIM. Results There were 77 PCS (44 boys) with a mean age of 10.76 ± 4.49 years. The mean value of fatigue scales were VAS 2.48 ± 1.48, Borg 0.61 ± 1.00 and MFS total score 71.14 ± 18.74. There was a statistically signicant negative relationship between age and VAS scores of PCS (p = 0.047). The girls were found more tired than boys (VAS; p = 0.031). A positive correlation was established between the type of tumor and MFS total score (p = 0.048). WeeFIM total result was 118.64 ± 11.37. While there was a statisticially signicant relationship between VAS Fatigue Score and WeeFIM total (p = 0.07), there was no such correlation between MFS total and WeeFIM total (p = 0.080) and Borg Fatigue Scale and WeeFIM total (p = 0.449). It is established that there is a relationship between level of fatigue and DLA. Risk factors like tumor type or treatment modalities for fatigue should be investigated in larger samples of specic survivor groups. Fatigue and its association with DLA should be screened clinically as a routine surveillence in PCS and treatment options be planned. was carried out with a mean age of 3.5 years after the end of the treatment. According to the body mass index (BMI) results of PCS, there was a positive correlation between BMI with VAS scores (r=-0.397; p = 0.049), and negative correlation between BMI with BORG scores (r = 0.322; p = 0.048). The distribution of the types of cancer in decreasing frequency was, n = 32 (41.6%) with the diagnosis of central nervous system tumor (CNS), 27 (35.1%) acute lymphoblastic leukemia (ALL), 6 (7.8%) lymphoma, 5 (6.5%) soft tissue sarcoma, 3 (3.9%) neuroblastoma, 3 (3.9%) Wilms tumor, and 1 (1.3%) bone tumor. There was a positive correlation between the type of tumor and MFS total score (r = 0.322; p = 0.048). Fourty-two (54.5%) of them had either chemotherapy (CT), or radiotherapy (RT) or surgery. The average length of cancer treatment was 28.35 ± 11.27 (min-max = 2–53) months. CT cycles of 7.27 (SD, 3.66), and RT sessions of 31.27 (SD, 9.06) were received. Thirty of the families (38.9%) were consanguineous marriages. Considering the educational status, 43 (55.8%) of the mothers and 27 (35.1%) of the fathers were primary school graduates.


Introduction
Fatigue is a distressing symptom which is one of the most common complaints of children with cancer [1]. High rates of physical impairment and lower physical functioning are the possible risk factors for fatigue. However, no de nite reasons are established [2]. Indeed, pediatric and young adult cancer survivors suffer from cancer-related fatigue even years after completion of treatment due to mostly fear of recurrence and other reasons [3]. Parents and health care providers of pediatric cancer survivors (PCS) report that sleep changes, anxiety and side effects of medications used during cancer treatment might also cause fatigue [4]. The national comprehensive cancer network of USA de ned cancer-related fatigue as "a distressing, persistent, subjective sense of physical, emotional and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity and interferes with usual functioning" [5]. Therefore fatigue often might probably have detrimental effects on daily living activities (DLA). It has negative impacts on many aspects such as personal relationships and school/ work in PCS [6]. Fatigue, poor physical status and impaired functional mobility are among the most commonly reported long-term sequela of pediatric cancer treatment [7]. Emotional distress, pain, and physical functioning limitations were all associated with increased level of fatigue in another survivorship study [8]. Decreased physical tness as a result of sequela of cancer treatment can cause DLA [4]. Many scales with symptom-related items have been developed to assess DLA in PCS [9].
Our primary goal was to evaluate cancer-related fatigue and its association with different types of measurements and second goal was to investigate the effect of fatigue on DLA in PCS. This study addresses a knowledge gap in the literature by exploring the level of impact of fatigue on DLA in PCS.

Methods
A quasiexperimental design was used to evaluate the feasibility and e cacy of fatigue on DLA in PCS. This study was approved by an Institutional Ethical Commity review board and research council at the a liated clinical site.

Participants
Participants were recruited consecutively as part of the assessment at a tertiary pediatric hematology and oncology clinic, between 2018 and 2020 in a regular outpatient clinics examination. Before referral to the unit, patients had been throughly examined by their primary care physician in order to exclude any other causes for the symptoms. After referral, consent forms and questionnaires were applied to the patients when they visit the clinic. In total 77 patients completed both consent forms and questionnaires. The children who were referred from pediatric hematology and oncology to physiotherapy rehabilitation department were included in the study. Background variables: Children background information included demographical data such as age, gender, and age at diagnosis and clinical characteristics of the illness.
PedsQL Multidimensional Fatigue Scale: The 18-item PedsQL TM Multidimensional Fatigue Scale (MFS) re ects three subscales: general fatigue, sleep/rest fatigue, and cognitive fatigue. The participants rated how often a particular problem occurred in the past month, using a 5-point Likert scale and for the young child self-report a 3-point scale. Each item is reverse-scored and rescaled to 0-100 scale, so that higher scores indicate fewer symptoms of fatigue [10].
Visual Analog Scale: The visual analog scale (VAS) is a valid and reliable measure, a tool with a 10-cm ruler and a marker that the patient moves to the point indicating his or her intensity were asked to rate pain intensity by placing a mark on a 100-mm VAS. The VAS was horizontally positioned with the extremes labeled "0= least possible fatigue" and "10= worst possible fatigue" [11].
Borg Fatigue Scale: Fatigue perception were determined using a 10 point modi ed Borg Scale during the walking test. Before the test at rest, the results of Perception of Dyspnea (Modi ed Borg Scale 0-10) and Perception of Fatique (Modi ed Borg Scale 0-10) value [12]. High scores indicate high level fatigue. The baseline values were used in our study.
WeeFIM: The Functional Independence Measure (FIM), builds on the organizational format of the Functional Independence Measure for Adults of the UDS for Medical Rehabilitation developed by Granger and colleagues, and the WeeFIM was modi ed for the children. The WeeFIM consists of 18 items involving six domains (self care, sphincter control, transfers, locomotion, communication, social cognition) for the children. In scoring the WeeFIM, a seven-level ordinal scale is used. Levels 6 and 7 re ect independence. No personal assistance is required by the child to successfully complete the task. However, level 6, modi ed independence, re ects the child's need for an assistive device (e.g., brace, wheelchair, grab bar), additional time to complete task, or concerns about safety in order to complete the task. Levels 3, 4, and 5 require modi ed dependence; either supervision or a degree of personal assistance is required to complete the task. Levels 2 and 1 indicate complete dependence; a score of 2 indicates maximal assistance (i.e., the child's need for the adult to perform the majority of the task); a score of 1 is given when the caretaker performs almost all the task for the child. The total score is the sum of all items. The motor total score was consisted of self care, sphincter control, transfers, and locomotion. The cognitive total score was included communication and social cognition [13].
Statistical Analysis SPSS software version 23.0 was used to analyze data. Student's t-test was used for discrimination analysis between groups after testing the ratio of the variances with an f-test and correcting the t-score according to the Smith-Satterthwaite method where appropriate. A con dence level of p<0.05 was considered statistically signi cant for reaching a statistical decision. Possible dependencies between outcome measures were investigated by computing Pearson's linear correlation coe cient "r". Detected correlations were deemed as "strong" if r≥ 0.70, "moderate" if 0.30 ≤r<0.70, and "weak" if r>0.30.
Correlation con dence levels were determined with one-tailed Student's t-test of the r 2 value.

Sample characteristics
In this cross-sectional chart review study, data were analyzed regarding 77 PCS. Demographic characteristics were presented in Table 1. Their mean age was 10.76 (min-max = 6-18) years with 44 (57.1%) boys. The correlation analysis of demographic and clinical characteristics were given in Table 2. There was a statistically signi cant negative relationship between age and VAS scores of PCS (r=-0.288; p = 0.047). Younger children were found to be more tired. The girls were found more tired than boys (VAS; r=-0.433; p = 0.031). The mean age of them at diagnosis was 7.14 (SD, 4.20) years. The evalution was carried out with a mean age of 3.5 years after the end of the treatment. According to the body mass index (BMI) results of PCS, there was a positive correlation between BMI with VAS scores (r=-0.397; p = 0.049), and negative correlation between BMI with BORG scores (r = 0.322; p = 0.048). The distribution of the types of cancer in decreasing frequency was, n = 32 (41.6%) with the diagnosis of central nervous system tumor (CNS), 27 (35.1%) acute lymphoblastic leukemia (ALL), 6 (7.8%) lymphoma, 5 (6.5%) soft tissue sarcoma, 3 (3.9%) neuroblastoma, 3 (3.9%) Wilms tumor, and 1 (1.3%) bone tumor. There was a positive correlation between the type of tumor and MFS total score (r = 0.322; p = 0.048). Fourty-two (54.5%) of them had either chemotherapy (CT), or radiotherapy (RT) or surgery. The average length of cancer treatment was 28.35 ± 11.27 (min-max = 2-53) months. CT cycles of 7.27 (SD, 3.66), and RT sessions of 31.27 (SD, 9.06) were received. Thirty of the families (38.9%) were consanguineous marriages. Considering the educational status, 43 (55.8%) of the mothers and 27 (35.1%) of the fathers were primary school graduates.    Table 1. The evaluation was carried out after 38 months in ALL group and 15 months in CNS group (p < 0.05) after the end of the treatment. CNS group was found to be signi cantly more tired than ALL group (VAS; CNS vs ALL = 3.27 vs 1.38; p = 0.000) (Fig. 1). According to Borg and MFS scores, there was no statistically signi cant difference between groups, but CNS group was more tired in terms of all scale scores. Some of WeeFIM scores in CNS group were signi cantly lower than the ALL group (Table 1). While there was a statistically signi cant difference between WeeFIM locomotion (p = 0.044), communication (p = 0.050), social cognition (p = 0.009) subheadings and cognitive total (p = 0.011) and total scores (p = 0.039), there was no statistically signi cant difference between self care (p = 0.039). =0.072), sphincter control (p = 0.620), transfers (p = 0.427) subheadings and motor total (p = 0.120) scores between CNS group and ALL group (Table 1 & Fig. 2).

Correlations
Hierarchical linear regression was conducted to determine the demographic and cancer related variables. Correlation between fatigue and DLA in PCS was given in Table 3. There was statistically signi cant correlation between VAS fatigue score and WeeFIM total scores (r=-0.387; p = 0.007) ( Table 3) in all study sample. According to our results, the important risk factors affecting fatigue were; age, gender, BMI, tumor type, time after completion of treatment, CT and RT. Younger age was found to be associated with higher level of fatigue. We cannot de nitely nd out the reason, but this is probably due to presence of large sample size of children with CNS tumor in whom their ages were younger than the leukemia group; becouse it is established that patients with CNS tumors were more tired than ALL group in our study. We did not nd an association between age at diagnosis and fatigue in our study like that of the [14][15][16]. Girls had high fatigue scores in our study. This is also probably related to the larger sample of CNS group in our study. This demographic nding was consistent with other various studies [14][15][16][17][18]. Higher BMI was associated with higher levels of fatigue in our study. Although Karimi et al. [4], having a sample of recent survivors like that of us did not report such association, Johnson et al. [19] reported clinically signi cant relationship with fatigue and weight. This study included brain tumor survivors of ages 8-12 years who were less than 6 years posttreatment, that is like our brain tumor study group. Studies with adult survivors of childhood cancers revealed an increased risk of fatigue with higher BMI [7,14]. Time after completion of treatment was 28.35 ± 11.27 months in our study. Our study time point was relatively an earlier time after completion of treatment than that of the literature, but our ndings are in concordant with Johnson et al. [19], Karimi et al. [4], and Nagai et al [17]. in which those studies are also the studies performed just recent years after treatment. Fatigue scores decreased with time elapsed in those studies. According to our study, in terms of CNS vs ALL group, probably because ALL group was older in age and a longer period of time had passed since the end of the treatment, a lower level of fatigue was detected in that ALL group. The average time since diagnosis was under 10 years in our study and those aformentioned studies [4,17,19]. This is a shorter time than that of Langeveld et al. [15], Meeske et al. [7] and Mulroney et al [14] in which fatigue levels declined over time during the long follow up of survivors. Tumor type, in our study, which indicates high number of patients with CNS tumor (40% of total) was found to be associated with higher fatigue scores. Johnson et al. [19], pointed the signi cance of fatigue in their brain study group of 21 patients. However, no signi cant effect was found between fatigue and primary cancer diagnosis in the literature. [4,[14][15][16][17][18]. Treatment modalities of both CT and RT were found to be signi cantly associated with higher levels of fatigue in this study. This might be probably due to large sample size of patients with CNS tumors in our study in whom all of them were treated with both CT and cranial RT. Johnson et al. [19] established a clinically signi cant relationship between fatigue and CT and cranial RT. Forty-two of 142 (30%) CNS tumor survivors were identi ed as having clinically signi cant fatigue in the study of "independent-complete independence" level. This shows that their levels are below the levels of healthy normal children [26]. It was seen that not only physical but also mental fatigue of children was re ected in their DLA. Patients with higher VAS scores (high fatigue score) did low performance of WeeFim self-care, social cognition, motor control, cognitive and total control. Patients with high MFS scores (low fatigue scores) had high performance in locomotion, social cognition and cognitive subscale. We found higher scores of fatigue in CNS tumor group than ALL group, in our study, and there was a signi cant association in terms of WeeFIM locomotion, communication, cognition, but not with sphincter control, transfers and motor total between these groups. CNS tumor group, with younger ages and having high risk of neurocognitive sequela due to surgery, cranial RT and CT, showed low WeeFim scores on communication and cognition. Hence, fatigue has not only physical impact but also neurocognitive adverse effect. DLA are more hampered in patients with CNS tumor in terms of self-care, socialization and cognition rather than locomotion. These results indicate that symptom pro les may be important to understand the variation between individuals in terms of impact of symptoms on functioning.
Our study has several clinical implications and limitations. First, PCS attending pediatric oncology centers should be screened for fatigue and would bene t from routine assessment of fatigue. Risk factors according to each patient should be identi ed and possible physical, social and cognitive impact should be screened by way of measurements of DLA. There should be interventions and solutions in management of symptoms of fatigue after ruling out other medical reasons. Given our study on self-reported data, information about past/current symptoms and other medical information may have been affected by recall bias. This study is cross-sectional with a wide age range of patients with heterogeinity. However, we suggest survivors with CNS tumor needs more attention and care in terms of fatigue and its outcome.
Future cohort studies with speci c and detailed evaluations are needed to improve surveillance and/or screening for fatigue, and mobility and of potential relevances like the maintenance of overweight. Future studies may bene t from assessments of symptomatic treatments according to tumor types. Christen et al. [3] mentioned that interventions that are useful are physical activity (level B evidence), education about cancer-related fatigue (level B evidence), relaxation and mindfullness (level C evidence, existing guidelines), cognitive behavioral therapy (level C evidence, existing guidelines), and adventure-based training (level C evidence). Growing evidence advocates physical activity interventions as bene cial, as well as feasible and safe, in managing cancer-related fatigue, and also improve quality of life and functional status during and after treatment in PCS [27].

Conclusion
In conclusion, fatigue is frequently experienced as a late effect of PCS. Particularly survivors with CNS tumor needs attention and follow-up. The study suggests that clinical physiotherapy rehabilitation evaluation might be successfully used in PCS in determining and treatment of fatigue.

Declarations
Funding: N/A Con ict of interest/Competing interests: No con ict of interest.
Availability of data and material: Dataset can be made available from the authors pending appropriate authorisations. The evaluation was carried out after 38 months in ALL group and 15 months in CNS group (p<0.05) after the end of the treatment. CNS group was found to be signi cantly more tired than ALL group (VAS; CNS vs ALL= 3.27 vs 1.38; p=0.000) (Figure 1).