Review | Published:

Is physical fitness decreased in survivors of childhood leukemia? A systematic review

Leukemia volume 19, pages 1317 (2005) | Download Citation



The aim of this review is to determine whether physical fitness, assessed by peak oxygen uptake (VO2peak) measurement, is reduced in survivors of acute lymphoblastic leukemia (ALL) compared to healthy children. A systematic literature search (up to June 2004) was performed using Medline, Sportdiscus, Cinahl, Embase, Cochrane and PEDro database and reference tracking. The VO2peak (ml kg−1 min−1) reached during a maximal exercise test until volitional exhaustion was used as the main outcome for this review. In all, 17 studies were identified in the literature. Data from three studies (102 ALL survivors, age ranging from 7 to 19 years) were pooled in a meta-analysis. Although there was a significant heterogeneity between the included studies (P=0.0006), the standardized mean difference (SMD) value of −0.61 (P=0.07) indicated that VO2peak tended to be reduced in survivors of childhood ALL compared to healthy control subjects, that is, decrease of −5.97 ml kg−1 min−1 (95% confidence interval (CI): (−12.35, 0.41); P=0.07) or −13% (95 % CI: (−27, 0.004)). Physical fitness tends to be reduced in survivors of ALL during childhood, which suggests the need for this population group to engage in regular physical activities with the purpose of increasing their functional capacity. Although more research is needed, this functional improvement might ameliorate the quality of life of ALL survivors as physical and outdoors activities are an essential part of daily routine during childhood.


The peak oxygen uptake (VO2peak) attained during a graded maximal exercise to volitional exhaustion is considered by the World Health Organization as the single best indicator of aerobic physical fitness.1 This variable, commonly expressed as the volume of oxygen consumed per unit of time relative to body mass (ml kg−1 min−1), is also a valid indicator of health status2 and a powerful predictor of mortality in both healthy and diseased individuals.3, 4 The improvements in exercise capacity and VO2peak brought about by training are related to improved quality of life (QOL), particularly in patients with exercise capacity limited by various disease processes.5

In adult cancer patients/survivors, it is not untypical to measure VO2peak levels considerably lower (50%) than predicted, which reflects the sedentary life habits and poor physical condition of this population group.6 Poor physical condition self-perpetuated by sedentarism is largely responsible for the disrupting symptoms of fatigue that these individuals experience during normal activities of daily living, with subsequent impairment in QOL.6 In turn, increases in functional capacity brought about by regular exercise training are reflected by higher VO2peak levels and result in improved OQL, that is, normal activities can be carried out with no fatigue.7, 8, 9, 10

Less research has focused on the physical capacity and VO2peak of children with cancer or survivors of cancer during childhood. Most studies have been performed on children survivors of acute lymphoblastic leukemia (ALL) and it is unclear whether their VO2peak is significantly decreased compared to healthy controls. For instance, Vizinova et al11showed no significant difference between ALL survivors and controls, while other authors found significantly decreased VO2peak levels in the former.12, 13, 14 Outdoor physical activities involving cardiorespiratory work of moderate intensity are an essential part of the daily routine of children.6 Thus, it would be interesting to assess if functional capacity, assessed by VO2peak measurement, is significantly decreased in children survivors of ALL. If this is the case, exercise training prescription is necessary to improve their QOL and more research is warranted in this field.

It was therefore our purpose to determine whether the physical fitness of ALL survivors, assessed with VO2peak, is decreased compared to healthy age-matched children.

Materials and methods

Search strategy

Publications were selected based on a literature search from 1966 until June 2004 using the Medline, Pubmed, Sportdiscus, Cinahl, Embase, Cochrane, and PEDRO database. Search terms ‘physical fitness’, ‘exercise testing’, ‘exercise’, ‘exercise capacity’, ‘exercise tolerance’, ‘child’, ‘survivors’, ‘acute lymphoblastic leukemia’, and ‘leukemia’ were used. References of the selected papers were tracked to find additional publications on this subject.

Selection of publications and types of outcome measures

We first selected all publications that reported one or more of the following outcome variables in ALL survivors: VO2peak (in ml kg−1 min−1), maximal heart rate, respiratory exchange ratio (RER), and exercise testing on a treadmill or cycle ergometer. Thereafter, we included in this study only those publications reporting: (1) data of survivors of childhood ALL included within the same homogeneous group (ie, excluding survivors of any other type of cancer during childhood) and their corresponding healthy controls, (2) VO2peak values (in ml kg−1 min−1) measured during a graded maximal exercise to volitional exhaustion, (3) description of methodology (gas-exchange analysis) for VO2peak measurement, and (4) description of subjects' characteristics (both ALL survivors and controls).

Data were extracted from the publications by two independent reviewers and entered into Review Manager 4.2.3 (Update Software, Oxford, UK).


DerSimonian and Laird Random Effects Model were used for analyzing the results on VO2peak because of the significant heterogeneity between the studies. The data were pooled using standardized mean differences (SMDs). SMD is the difference between two means divided by an estimate of the within-group standard deviation, and can be considered as an effect size, for example, negative values for SMD would indicate a lower physical fitness of childhood survivors of ALL compared to healthy controls. The level of statistical significance was set at P<0.05.


A total of 17 published studies (two of which were written in the Czech11 and Polish language15) were identified in the literature. Of them, 14 did not meet the criteria described above for inclusion in the meta-analysis (Table 1). The study by Matthys et al13 was excluded because they included in the same group ALL survivors and childhood survivors of several types of cancer other than ALL. The studies from Ostanski et al,15 Black et al,16 Kadota et al,17 Lipshultz et al,18 Pihkala et al,19 Calzolari et al20 Turner-Gomez et al,21 and Zalewska-Szewczyk et al22 were excluded from the meta-analysis, because of missing or incomplete description of the methodology (gas analysis) for VO2peak measurement. The study by Prestor et al23 lacked a clear description of the patients. Johnson et al24 reported only the results of submaximal testing and did not report VO2peak values directly measured during a maximal graded test to exhaustion. Jenney et al25 expressed the results as percent predicted values, and did not report absolute values of VO2peak. Sharkey et al26 did not describe their control subjects, and McKenzie et al29 studied only patients treated for solid tumor cancers.

Table 1: Studies excluded from the meta-analysis

A total of 102 childhood survivors of ALL and 99 control subjects from three studies were included in this review (Table 2). For the study that reported VO2peak values separately for girls and boys,14 VO2peak data were also entered separately into the meta-analysis. One study reported values separately for patients with and without normal stress echocardiography.12 The VO2peak data for this study were also entered separately into the present meta-analysis. All included studies used a calibrated metabolic cart for gas-exchange analysis. Despite variations between studies in exercise mode (subjects pedaling on a cycle ergometer11 or running/walking on a treadmill12, 14) and protocols (ie, different rates of workload increases to attain exhaustion) and instrumentation (ie, different commercial models of metabolic carts for gas-exchange measurement), the measurement of VO2peak values was based on the same methodology in all the three included studies. None of the studies included in the meta-analysis specified the number of patients who underwent bone marrow transplantation (BMT). This is to be kept in mind as BMT is associated with reduced VO2peak levels in survivors of childhood ALL.25

Table 2: Studies included in the meta-analysis

The results of the meta-analysis are displayed in Table 3. Although there was a significant heterogeneity between the included studies (P=0.0006), the SMD value of −0.61 (P=0.07) indicated that VO2peak tends to be reduced in survivors of childhood ALL compared to healthy control subjects, that is, decrease of −5.97 ml kg−1 min−1 (95% confidence interval (CI): (−12.35, 0.41)) or −13% (95% CI: (−27, 0.004)).

Table 3: Forrest plot with the comparison of VO2peak values of ALL patients with controls


The results of the present systematic review indicate that the VO2peak values (and thus the fitness level) of survivors of childhood ALL tend to be reduced (average of −6 ml kg−1 min−1 or −13%) compared to healthy controls. Besides the decrease in QOL associated with lower VO2peak levels, the aforementioned average decrease in the VO2peak of ALL survivors is of clinical relevance as this variable is a powerful predictor of mortality in both healthy and diseased individuals,3, 4 for example, a −3.5 ml kg−1 min−1 reduction is associated with a 12% decrease in the survival rates of diseased people.4

There existed some heterogeneity between studies, mainly attributable to the study of Vizinova et al.11 These authors did not find significant differences between ALL survivors and control subjects. It must be, however, noticed that ALL survivors were encouraged to be physically active, whereas the children of the control group followed a sedentary lifestyle. The second factor that contributed to the heterogeneity between studies arises from the inclusion of ALL survivors with normal stress echocardiography in one of the studies.12 Indeed, Hauser et al12 found no significant differences in exercise capacity between ALL survivors with normal stress echocardiography results and healthy control children. This finding suggests that impaired cardiac function is responsible, at least partly, for the reduced functional capacity of ALL survivors, as discussed below. On the other hand, the wide CI we obtained (impairment −13%, 95% CI: (−27, 0.004)) indicates a considerable variation in the physical fitness levels of ALL patients after successful treatment. This might be due to differences in treatment and response to treatment in leukemia patients. For instance, Sharkey et al26 found normal exercise capacity in patients receiving minimal doses of anthracycline and no irradiation.

In humans, decreases in VO2peak are largely attributable to impaired cardiac function as VO2peak mainly reflects (and is largely limited by) maximal O2 supply to muscles (ie, maximal cardiac pump capacity) rather than maximal rate of O2 utilization by muscle mitochondria.27, 28 Anticancer therapy may affect central cardiac dynamics and thus blood supply to body tissues, particularly exercising muscles. Anthracyclines can induce myocardial damage (eg, doxorubicin-induced cardiomyopathy) with subsequent decreases in cardiac output.29, 30 Sedentary habits (especially bed rest) induce cardiac atrophy and further reduce stroke volume and thus cardiac output in young adults31 and children.32 Since sedentary ALL survivors with reduced VO2peak are able to reach normal values of maximal heart rate during exercise,13, 14 impaired stroke volume is largely responsible for their reduced cardiac output and thus decreased VO2peak (as cardiac output is the product of heart rate by stroke volume). Anticancer therapy can also alter the exercise capacity of ALL survivors due to its deleterious effects on lung function. Craniospinal irradiation, cyclophosphamide or lung infections during or subsequent to treatment for leukemia (eg, bacterial, or due to respiratory syncytial virus, candida, pneumocystis or cytomegalovirus) can reduce total lung capacity.25 Lung function impairment in ALL survivors is reflected by the occurrence of arterial desaturation (oxygen saturation values <90%) during exercise.13

Besides insufficient pumping of oxygenated blood to working muscles, several phenomena at the peripheral (muscle tissue) level might severely limit the maximal capacity of muscle fibers to consume oxygen and further decrease the VO2peak of ALL survivors. Muscle atrophy is a common problem in this population group due to the catabolic effects of several chemotherapeutic agents as vincristine or corticosteroids.14, 33 Muscle atrophy implies a smaller muscle mass to consume oxygen during exercise. In addition, the metabolic function of muscle fibers can be altered. Impaired aerobic metabolism (due to decreased mitochondrial volume and/or mitochondrial myopathy) or reduced capillarization can occur after immunosuppressive therapy.34 Muscle atrophy and altered muscle function are further aggravated by sedentary habits due to the catabolic effects that sedentarism and prolonged bed rest induce on skeletal muscle tissue.6 As a result, muscle atrophy and early fatigue during low-to-moderate physical tasks become self-perpetuating conditions.6 Although more research is needed, some data suggest that the total daily energy expenditure of ALL survivors is indeed reduced compared to healthy children, leading to further deconditioning. Particularly, female patients show a greater impairment after treatment with anthracyclines, which results in further deterioration of functional capacity and increased body fat compared to males.13, 14, 16 For instance, Matthys et al estimated the physical activity of ALL survivors with a questionnaire. Sport leisure time was lower in girl survivors of cancer (the majority, but not all, of which was ALL) than in their corresponding controls. Their VO2peak levels were also lower (−8 ml kg−1 min−1).13 Warner et al35 measured total daily energy expenditure (TDEE) and physical activity levels (=TDEE/basal metabolic rate) in long-term survivors of ALL and compared them with results from survivors of other malignancies and healthy sibling control subjects. The median TDEE was reduced in the ALL group (150 kJ kg day−1) compared with other malignancies and controls (207 and 185 kJ kg day−1, respectively). In turn, this reduction was accounted for mainly by a relative decrease in their levels of physical activity. Total energy expenditure and physical activity were in turn correlated with percentage body fat, indicating that obesity in survivors of ALL may, in part, be explained by a decrease in their TDEE as a consequence of their low physical activity levels. The detrimental effects of sedentarism are aggravated by the fact that diseased children may underestimate their own potential for performing physical tasks due to low self-esteem or overprotection by their parents.13 Only physical training can break the ‘vicious circle’ of sedentary habits and subsequent exercise intolerance.6

Finally, it must be kept in mind that glucocortocoid therapy can increase adiposity and body mass in children receiving treatment for ALL.36 As VO2peak is expressed relative to body mass (ml kg−1 min−1), the decreased VO2peak levels of ALL survivors might be partly attributable to the aforementioned side effects of glucocorticoid treatment. Nevertheless, the difference in VO2peak levels between the ALL survivors and controls included in our meta-analysis cannot be fully explained by differences in body mass, as the mean values of this variable were very similar in both groups of subjects studied by Vizinova et al11 (ALL survivors: 46.2 kg; controls: 45.1 kg) and Warner et al14 (boys survivors of LLA: 50.9 kg; controls: 51.0 kg), except for the girls included in the report by Warner et al14 (ALL survivors: 49.0 kg; controls: 45.5 kg). Although Hauser et al12 did not report body mass, they stated that their controls and ALL survivors were ‘matched for age and body surface area’.

Conclusions and perspectives

The physical fitness (as reflected by VO2peak levels) of ALL survivors tends to be reduced compared to healthy children. Impaired physical fitness leads to early fatigue during physical activities and can severely deteriorate the QOL of ALL survivors, which suggests the need for these children to engage in regular physical activities. Exercise physiologists could assist oncologists in prescribing exercise programs for attenuating cancer-related fatigue and help improve the physical fitness and QOL of children surviving cancer.6 Furthermore, there are scientific indications that exercise training improves the function of several anti-cancer immune system components,37 and can attenuate tumor development.38


  1. 1.

    , , , , , et al. The maximum oxygen intake. An international reference standard of cardiorespiratory fitness. Bull World Health Organ 1968; 38: 757–764.

  2. 2.

    , , (eds). Exercise Physiology. Energy, Nutrition and Performance, 5th edn. Philadelphia, PA: Lippincott Williams & Wilkins, 2001, pp 174–186.

  3. 3.

    , , , , , et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA 1996; 276: 205–210.

  4. 4.

    , , , , , . Exercise capacity and mortality among men referred for exercise testing. N Engl J Med 2002; 4: 793–801.

  5. 5.

    , , , , , et al. Physical activity and exercise training prescriptions for patients. Cardiol Clin 2001; 19: 447–457.

  6. 6.

    , , . Cancer-related fatigue: can exercise physiology assist oncologists? Lancet Oncol 2003; 4: 616–625.

  7. 7.

    , , , , . Effects of physical activity on the fatigue and psychologic status of cancer patients during chemotherapy. Cancer 1999; 85: 2273–2277.

  8. 8.

    , , . Effects of aerobic interval training on cancer patients' functional capacity. Nurs Res 1989; 38: 348–351.

  9. 9.

    , . Effects of exercise on physiological and psychological variables in cancer survivors. Med Sci Sports Exerc 2002; 34: 1863–1867.

  10. 10.

    , , , . Exercise reduces fatigue in chronically fatigued Hodgkins disease survivors – results from a pilot study. Eur J Cancer 2003; 39: 57–63.

  11. 11.

    , , , . Exercise cardiorespiratory indexes in children motivated to physical activity after treatment for acute lymphoblastic leukemia. Cas Lek Cesk 2002; 141: 491–493.

  12. 12.

    , , . Diagnosis of anthracycline-induced late cardiomyopathy by exercise-spiroergometry and stress-echocardiography. Eur J Pediatr 2001; 160: 607–610.

  13. 13.

    , , , , , . Gender difference in aerobic capacity in adolescents after cure from malignant disease in childhood. Acta Paediatr 1993; 82: 459–462.

  14. 14.

    , , , . Relationship between cardiopulmonary response to exercise and adiposity in survivors of childhood malignancy. Arch Dis Child 1997; 76: 298–303.

  15. 15.

    , . Exercise tolerance in patients after acute lymphoblastic leukemia treatment in childhood. Wiad Lek 2001; 54: 650–655.

  16. 16.

    , , . Physical performance in long-term survivors of acute leukaemia in childhood. Eur J Pediatr 1998; 157: 464–467.

  17. 17.

    , , , , . Cardiopulmonary function in long-term survivors of childhood Hodgkin's lymphoma: a pilot study. Mayo Clin Proc 1988; 63: 362–367.

  18. 18.

    , , , , , . Late cardiac effects of doxorubicin therapy for acute lymphoblastic leukemia in childhood. N Engl J Med 1991; 324: 808–815.

  19. 19.

    , , , , , et al. Cardiopulmonary evaluation of exercise tolerance after chest irradiation and anticancer chemotherapy in children and adolescents. Pediatrics 1995; 95: 722–726.

  20. 20.

    , , , , , et al. Evaluation of a group of leukaemic children ‘off-therapy’, towards their inclusion in physical activities. Int J Sports Cardiol 1985; 2: 108–115.

  21. 21.

    , , , , , et al. Cardiorespiratory status after treatment for acute lymphoblastic leukemia. Med Pediatr Oncol 1996; 26: 160–165.

  22. 22.

    , , . Late cardiotoxicity of anthracyclines in children with acute leukemia. Klin Padiatr 1999; 211: 356–359.

  23. 23.

    , , , , , . Cardiovascular responses to dynamic submaximal exercise in children previously treated with anthracycline. Am Heart J 1997; 133: 169–173.

  24. 24.

    , , , . Late cardiac damage of anthracycline therapy for acute lymphoblastic leukemia in childhood. Pediatr Hematol Oncol 2000; 17: 527–540.

  25. 25.

    , , , . Lung function and exercise capacity in survivors of childhood leukaemia. Med Pediatr Oncol 1995; 24: 222–230.

  26. 26.

    , , , . Cardiac rehabilitation after cancer therapy in children and young adults. Am J Cardiol 1993; 71: 1488–1490.

  27. 27.

    , . Limiting factors for maximum oxygen uptake and determinants of endurance performance. Med Sci Sports Exerc 2000; 32: 70–84.

  28. 28.

    , , , , , et al. Evidence of O2 supply-dependent VO2max in the exercise-trained human quadriceps. J Appl Physiol 1999; 86: 1048–1053.

  29. 29.

    , , , , . Aerobic and anaerobic capacities of children and adolescents successfully treated for solid tumors. Clin Exerc Physiol 2000; 2: 39–42.

  30. 30.

    . Effects of exercise on cancer-related fatigue. Cancer 2001; 92 (Suppl): 1689–1693.

  31. 31.

    , , . Cardiac atrophy after bed-rest deconditioning: a nonneural mechanism for orthostatic intolerance. Circulation 1997; 96: 517–525.

  32. 32.

    , , , , , . Long-term effect of previous swim training in girls. A 10-year follow-up of the ‘girl swimmers’. Acta Paediatr Scand 1978; 67: 285–292.

  33. 33.

    , , , . Impaired muscle strength in female adolescents and young adults surviving leukemia in childhood. Cancer 1993; 72: 276–281.

  34. 34.

    , . Exercise and inhibition of glucocorticoid-induced muscle atropy. Exerc Sport Sci Rev 1993; 21: 135–167.

  35. 35.

    , , , . Daily energy expenditure and physical activity in survivors of childhood malignancy. Pediatr Res 1998; 43: 607–613.

  36. 36.

    , , , , . Short-term effects of prednisolone and dexamethasone on circulating concentrations of leptin and sex hormone-binding globulin in children treated for acute lymphoblastic leukaemia. Clin Endocrinol 2003; 58: 770–776.

  37. 37.

    , , , . Physical exercise and immune system function in cancer survivors: a comprehensive review and future directions. Cancer 2002; 94: 539–551.

  38. 38.

    , , . Attenuation of the development of murine solid leukemia tumor by physical exercise. Antioxid Redox Signal 2002; 4: 213–219.

Download references


This study was financed by a grant from the ‘Stichting Nationaal Fonds tegen Kanker’, The Netherlands. Alejandro Lucia is supported by Strive Inc (USA).

Author information


  1. Departments of Pediatric Physical Therapy & Exercise Physiology, University Hospital for Children and Youth ‘Het Wilhelmina Kinderziekenhuis’, University Medical Center Utrecht, Utrecht, The Netherlands

    • M van Brussel
    • , T Takken
    • , J van der Net
    •  & P J M Helders
  2. Physiology Department, European University of Madrid, Spain

    • A Lucia


  1. Search for M van Brussel in:

  2. Search for T Takken in:

  3. Search for A Lucia in:

  4. Search for J van der Net in:

  5. Search for P J M Helders in:

Corresponding author

Correspondence to T Takken.

About this article

Publication history