Effects of supervised cardiovascular training program on exercise tolerance, aerobic capacity, and quality of life in patients with systemic lupus erythematosus




To determine if supervised cardiovascular training improves exercise tolerance, aerobic capacity, depression, functional capacity, and quality of life in patients with systemic lupus erythematosus (SLE).


Sixty women with SLE (ages 18–55 years) were evaluated using Short Form 36, visual analog scale for pain, scale for fatigue, Beck Depression Inventory, and Health Assessment Questionnaire (HAQ), and participated in a training protocol of incremental load on a treadmill with computed gas metabolic analysis. Maximum oxygen consumption (VO2max) and anaerobic threshold VO2 were calculated with a SensorMedics Vmax29C analyzer (Sensor Medics, Yorba Linda, CA), and heart rate was measured by electrocardiogram. Patients were divided into 2 groups: a training group (41 patients) that participated in the supervised cardiovascular training program and a control group (19 patients) that did not participate in the program. All variables were analyzed at baseline and after 12 weeks for both groups. The training program occurred in the morning for 60 minutes, 3 times a week for 12 weeks. Statistical analysis included Wilcoxon's rank sum test, Mann-Whitney U test, chi-square test, and Fisher's exact test. P values <0.05 were considered to be statistically significant.


The 2 groups were homogeneous and comparable at baseline. The training group showed a significant improvement of aerobic capacity measured by anaerobic threshold VO2 (14.67 ± 3.03 versus 17.08 ± 3.35 ml/kg/minute, P < 0.001). Comparison of the training group and control group after 12 weeks showed a significant difference relating to VO2max (24.31 ± 4.61 versus 21.21 ± 3.88 ml/kg/minute, P = 0.01) and anaerobic threshold VO2 (17.08 ± 3.35 versus 13.66 ± 2.82 ml/kg/minute, P < 0.0001). After cardiovascular training, we found a significant improvement of Beck inventory score (8.37 ± 12.79 versus 2.90 ± 3.00, P < 0.001) and HAQ score (0.14 ± 0.21 versus 0.06 ± 0.19, P < 0.01) in the training group.


This study showed significant improvement in exercise tolerance, aerobic capacity, quality of life, and depression after a supervised cardiovascular training program in patients with SLE.


Patients with systemic lupus erythematosus (SLE) can present limitations in exercise capacity (1) and reduced quality of life (2, 3) due to various clinical conditions such as fatigue (4), pulmonary disease (5, 6), heart disease, and peripheral neuropathy (7), and due to association with other rheumatic diseases such as fibromyalgia (8, 9). Previous studies have demonstrated that patients with SLE have lower aerobic capacity (10), lower oxygen pulse during exercise (11), greater fatigue (1), and worse quality of life (12) compared with normal controls. These patients may benefit from exercise, as observed in other populations, relating to immunologic aspects (13, 14), prevention of coronary disease (15, 16), and improvement in quality of life (17, 18).

There are few studies in the literature evaluating the effects of exercise on patients with SLE that demonstrate no significant improvement in aerobic capacity or quality of life and nevertheless demonstrate improvement in fatigue (1, 19, 20). The purpose of the present study was to evaluate if supervised cardiovascular exercise promotes improvement in exercise tolerance, aerobic capacity, fatigue, depression, and quality of life in women with SLE.


Women with SLE who were followed up at the rheumatology division of 2 university hospitals (Universidade Federal de São Paulo [UNIFESP] and Santa Casa de Misericórdia de São Paulo) were enrolled in this blind longitudinal study. A total of 350 patients followed up at UNIFESP and 72 followed up at Santa Casa were contacted during a medical visit. The majority of patients did not have free time to participate in the training program. Eligible patients who agreed to participate in the study were assigned to 2 groups: a training group that participated in a supervised training program for 12 weeks, and a control group that did not participate in training. For each patient in the training group, there was a patient in the control group who matched in age, body weight, height, and body mass index. After 12 weeks, patients in the control group were invited to participate in the training group, and 19 of those who agreed to participate were included in the total number of training group patients.

The groups were analyzed at baseline (T0) and after 12 weeks (TF). The training program, which took place for 60 minutes in the morning, consisted of 10 minutes of initial warmup/stretching, 40 minutes of walking, and 10 minutes of cooling down. The walk was performed at a heart rate corresponding to the ventilatory anaerobic threshold that was previously obtained in an ergospirometric test and monitored through a Polar frequency meter (Polar, Kempele, Finland) throughout the training period. Heart rate (HR) values were recorded at 10-minute intervals. The program was conducted 3 times a week for 12 consecutive weeks and was supervised by 2 physiotherapists on alternate days. The Research and Ethics Committee of UNIFESP approved the protocol, and all participants signed an informed consent form.

Patients were selected according to the following inclusion criteria: SLE according to the American College of Rheumatology classification criteria (21), age 18–55 years, and signed informed consent form. Exclusion criteria were as follows: hemoglobin values <10 gm/dl, neurologic disease or cardiovascular accident sequels, psychosis, a diagnosis of depression and/or patients under psychiatric care, presence of respiratory diseases (pulmonary hypertension, pulmonary fibrosis, bronchitis, asthma, emphysema), heart insufficiency (functional class ≥ II), a history of myocardial infarction or ischemic heart disease, diastolic blood pressure >100 mm Hg, active nephritis with creatinine levels >3.0 mg/dl, Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) scores >8, thyroid dysfunction, diabetes mellitus, hip and/or knee joint prosthesis or aseptic bone necrosis, deep venous thrombosis in the lower limbs, severe arthritis in ≥3 weight-bearing joints, pregnancy, and practicing regular physical activity (regular physical exercise ≥3 times a week for patients and controls) (22). Patients with SLE with concomitant rheumatic disease such as rheumatoid arthritis, systemic sclerosis, dermatomyositis, and fibromyalgia were also excluded.

Variables studied.

The influence of SLE on quality of life was evaluated using the Beck Depression Inventory questionnaire (BDI) (23), which is used to evaluate depression; the Health Assessment Questionnaire (HAQ) (24, 25), which is used to evaluate functional capacity; and the numerical visual analog scale for pain (26, 27), ranging from 0 to 10. The fatigue scale developed by Krupp et al (ranging from 1 to 7) was used to assess fatigue (28), and quality of life was assessed using the Short Form 36 (SF-36) (29, 30), a questionnaire consisting of 8 domains scored on a scale from 0 to 100, with a higher score indicating better function. A blind rater performed the evaluations the day of the cardiorespiratory test.

Physiologic variables.

Cardiorespiratory evaluation was performed using the physiologic exercise laboratory of the Universidade Federal de São Paulo (Centro de Medicina da Atividade Física e do Esporte, Sao Paulo, Brazil) in the morning, with room temperature between 19°C and 24°C. The equipment was calibrated with known volumes before each cardiorespiratory test (31) (3-liter Calibration Syringe, Series 5530; Hans Rudolph, Kansas City, MO). The gas analyzer was first calibrated with 2 cylinders, one with 26% O2 and the other with 4% CO2, 16% O2 (SensorMedics, Yorba Linda, CA), and also with ambiance air.

The cardiorespiratory tests were performed by a physiotherapist and a blinded biologist with an expert cardiologist. All participants were previously instructed to sleep for an adequate period and to avoid effort on the day before the test. On the day of the test, participants were instructed to have a meal 2 hours before the test; to abstain from stimulants such as tea, coffee, and soft drinks; and to wear comfortable clothing and shoes. Before beginning the cardiorespiratory test, patients and controls were weighed and measured for height, and were encouraged to perform the ergospirometric test up to maximal exercise capacity.

After a resting electrocardiogram (EKG) and arterial pressure measurement were obtained, patients underwent a protocol of incremental load on a computerized treadmill (SensorMedics 2.000 treadmill). After gas analysis for 3 minutes in the standing position, the participants performed an incremental treadmill test from 3.0 km/hour to exhaustion (after the first 3 minutes, the velocity was maintained at 4.0 km/hour for 2 minutes and progressively increased to 5.0 km/hour, 6.0 km/hour, and 7.0 km/hour, with each stage lasting 1 minute). After the first minute at 7.0 km/hour, the treadmill's angle of inclination was successively increased to 3%, 5%, 8%, and 10% at a constant velocity, with each inclination being maintained for 1 minute. The velocity was then increased to 8.0 km/hour with 10% and 15% inclination and finally increased to 9.0 km/hour with 15% inclination (each stage lasted 1 minute). After obtaining the maximum effort from each participant, the test was interrupted and the velocity was maintained for 2 minutes at 3.0 km/hour for recovery. Participants stood for an additional 6 minutes for gas analysis, EKG, arterial pressure measurement, and control of vital signs. The maximum time of exercise during the ergospirometric test was considered exercise tolerance and was measured at baseline and after 12 weeks.

Maximum oxygen consumption (VO2max) and anaerobic threshold VO2 were calculated with a SensorMedics Vmax29C series analyzer (VISION program). The mean computerized breath-by-breath gas metabolic analysis was calculated at 20-second intervals (32, 33) for the following variables: VO2 (ml/kg/minute − mean gas volume at standard temperature and pressure), ventilatory equivalent of oxygen (VE/VO2), carbon dioxide production (VECO2), pulmonary ventilation (liters/minute), expired O2 fraction, gas exchange ratio (R) (31), and oxygen pulse.

VO2max was considered to have been reached during the cardiorespiratory evaluation when ≥2 of the following criteria were satisfied: VO2 plateau/stabilization, R > 1.10, maximum HR reached (220 minus age), and physical exhaustion. For anaerobic threshold VO2, we considered the point of inflexion of the VE/VO2, and not the linear increase of expired gas (VE) and the anterior point at which the ratio between gas changes became greater than 1 (R > 1.0) due to loss of linearity between VECO2/VO2 (22, 31).

To determine the rate of perceived exertion during the test, the modified Borg scale (34), ranging from 0 to 10, was used. The scale was explained to participants before they started the test. During the test, immediately before changing the speed of the treadmill, the scale was shown to the participant on a clipboard, and the participant rated his or her effort to walk at that time. All participants were instructed to point at a number with their hand instead of talking.

HR was measured continuously throughout the protocol using a 12-lead EKG (35). A cardiologist analyzed the EKGs for the detection of ischemic alterations, and 2 expert blinded raters (a cardiologist and an exercise physiologist) analyzed VO2.

All variables were analyzed at T0 and TF in the 2 groups. The variables were also analyzed at T0 and at TF within each group of patients (intragroup comparison) and between groups (intergroup comparison).

Statistical analysis.

Data were analyzed using SPSS version 10.01 (SPSS, Chicago, IL). Student's t-test was used to assess the frequency of the qualitative variables for the 2 groups. Wilcoxon's rank sum test was used to analyze the variables measured at T0 and TF in the same group. The chi-square test and Fisher's exact test were used to analyze the percentage of improvement of anaerobic threshold VO2 and VO2max. The Mann-Whitney U test was used to compare quantitative variables between the training and control groups. P values less than 0.05 were considered statistically significant (36).


Sixty patients completed the study. Among the 50 patients assigned to the training group, 41 completed the cardiovascular training program and 9 dropped out for personal reasons. Of the 22 patients assigned to the control group, 19 were reevaluated after 12 weeks and 3 were not. At T0, patients in the training group and control group were homogeneous with respect to age, body weight, height, body mass index, hemoglobin (gm/dl), creatinine (mg/dl), SLEDAI score, and disease duration (Table 1). Racial distribution and the proportion of smokers were also similar between the 2 groups (Table 2). The medication used by all participants is shown in Table 3.

Table 1. Measurements of central tendency for all variables characterizing the groups at baseline of the study*
Baseline variablesTG (n = 41)CG (n = 19)P
  • *

    Values are the mean ± SD unless otherwise indicated. TG = training group; CG = control group; BMI = body mass index; Hgb = hemoglobin; SLEDAI = Systemic Lupus Erythematosus Disease Activity Index.

  • Mann-Whitney U test.

Age, years36.22 ± 10.7935.21 ± 9.130.725
Weight, kg66.46 ± 13.3463.62 ± 11.210.424
Height, cm158.01 ± 7.0157.53 ± 6.570.801
BMI, height2/kg26.64 ± 4.8625.81 ± 4.370.579
Hgb, gm/dl12.75 ± 1.3112.55 ± 1.350.579
Creatinine, mg/dl0.99 ± 0.660.84 ± 0.100.329
SLEDAI1.15 ± 2.011.75 ± 2.450.261
Disease duration, years5.84 ± 4.846.56 ± 3.570.426
Table 2. Proportion measurements for the variables between the training group (TG) and the control group (CG)*
VariablesTG (n = 41)CG (n = 19)P
  • *

    Values are the number (percentage). NS = not significant.

  • Chi-square test.

 White15 (36.6)8 (42.1)NS
 Nonwhite26 (63.4)11 (57.9) 
 No32 (80.0)15 (78.9)NS
 Yes3 (7.5)2 (10.5) 
 Former smoker25 (12.5)2 (10.5) 
Table 3. Medications taken by patients with systemic lupus erythematosus*
MedicationsTG (n = 41)CG (n = 19)
  • *

    Values are the number (percentage). TG = training group; CG = control group.

  • One patient from the training group was taking beta-blockers.

Prednisone, ≥5 mg/day14 (34)10 (53)
Hydroxychloroquine, 250 mg/day15 (27)9 (47)
Prednisone and hydroxychloroquine8 (20)5 (26)
Methotrexate, >7.5 mg/week3 (7)3 (6)
Antihypertensive drugs7 (17)1 (5)

Patients in the training group showed a significant improvement in aerobic capacity measured on the basis of anaerobic threshold VO2 when comparing T0 and TF values (mean ± SD 14.67 ± 3.03 ml/kg/minute versus 17.08 ± 3.35 ml/kg/minute; P < 0.001). Although at T0 the difference was not significant, training group patients showed a significantly higher VO2max at TF than the untrained patients (mean ± SD 24.31 ± 4.61 ml/kg/minute versus 21.21 ± 3.88 ml/kg/minute; P = 0.014). The trained patients also had a significantly higher anaerobic threshold VO2 than the controls at TF (17.08 ± 3.35 ml/kg/minute versus 13.66 ± 2.82 ml/kg/minute; P < 0.001) (Table 4). The control group showed no significant difference between the evaluations performed at T0 and TF, except for a worse anaerobic threshold VO2. Exercise tolerance and O2 pulse improved significantly after the training program in the training group (Table 4).

Table 4. Measurements of central tendency for physiologic variables characterizing the study groups at baseline (T0) and after 12 weeks (TF)*
VariablesTG (n = 41)CG (n = 19)
  • *

    Values are the mean ± SD unless otherwise indicated. TG = training group; CG = control group; VO2max = maximum oxygen consumption; AT = anaerobic threshold; HR = heart rate.

  • TG intragroup comparison, Wilcoxon's test.

  • CG intragroup comparison, Wilcoxon's test.

  • §

    Significant P values.

Maximum exercise tolerance, minutes10.46 ± 1.6311.93 ± 1.65< 0.001§10.91 ± 1.6411.11 ± 1.510.555
VO2max, ml/kg/minute22.63 ± 4.2524.31 ± 4.610.02§22.40 ± 4.6921.21 ± 3.880.164
AT, ml/kg/minute14.67 ± 3.0317.08 ± 3.350.001§15.27 ± 2.9113.66 ± 2.820.006§
Maximum ventilation, liters/minute63.75 ± 12.2471.01 ± 13.500.001§63.51 ± 15.8065.05 ± 13.380.441
Gas exchange ratio1.12 ± 0.091.14 ± 0.050.1921.14 ± 0.081.17 ± 0.060.314
Basal HR, beats/minute73.98 ± 8.9871.71 ± 9.220.12679.47 ± 75.6375.63 ± 10.100.050
AT HR, beats/minute132.76 ± 26.73140.12 ± 18.630.062130.06 ± 33.84125.74 ± 34.400.098
Maximum HR, beats/minute173.12 ± 16.45175.24 ± 15.060.169170.89 ± 14.87171.58 ± 14.110.673
Borg scale5.53 ± 2.746.61 ± 2.830.009§5.815 ± 3.146.16 ± 3.130.422
O2 pulse of VO2max, median ml/beats7.808.600.021§8.007.600.332
O2 pulse of AT VO2, median ml/beats6.107.300.001§7.006.400.061

HR and the gas exchange ratio did not differ between groups at T0 or TF, but a significant improvement in pulmonary ventilatory capacity was observed in the training group after the training program (mean ± SD 63.75 ± 12.24 versus 71.01 ± 13.50; P < 0.001) (Table 4).

No significant difference in the visual analog scale for pain was detected between T0 and TF in either group. However, a significant improvement in the depression scores measured by the BDI (mean ± SD score 8.37 ± 12.79 versus 2.90 ± 3.00; P < 0.001) and in the functional capacity scores assessed by the HAQ (0.14 ± 0.21 versus 0.06 ± 0.19; P < 0.010) was observed in the training group at the final evaluation. However, the comparisons between the training group and the control group at the end of the study for these 3 variables showed a significant difference only for functional capacity (HAQ), as shown in Table 5. The results of the Borg Scale variables are presented in Table 4, and the results of fatigue scores and the SF-36 domains for both groups are presented in Table 5.

Table 5. Measurements of pain, functional capacity, depression, and fatigue at baseline (T0) and after 12 weeks (TF)*
MeasureTG (n = 41)CG (n = 19)TG and CG P value§
  • *

    Values are the mean ± SD unless otherwise indicated. TG = training group; CG = control group; VAS = visual analog scale for pain; HAQ = Health Assessment Questionnaire; BDI = Beck Depression Inventory; SF-36 = Short Form 36.

  • TG intragroup comparison.

  • CG intragroup comparison.

  • §

    Intergroup comparison at T0 and TF.

  • Significant P values.

VAS2.02 ± 2.731.70 ± 2.690.472.47 ± 2.713.01 ± 3.440.720.520.10
HAQ0.14 ± 0.210.06 ± ± 0.270.38 ± 1.140.880.310.03
BDI8.37 ± 12.792.90 ± 3.00< 0.0015.79 ± 6.446.63 ± 8.500.890.350.15
Fatigue3.57 ± 1.472.68 ± 1.33< 0.0013.28 ± 1.333.29 ± 1.470.970.430.10
 Functional capacity86.34 ± 12.0491.10 ± 11.370.0187.89 ± 12.9486.84 ± 11.210.790.520.09
 Physical fitness70.12 ± 39.2285.24 ± 27.320.0368.42 ± 43.1760.53 ± 43.550.410.940.02
 Pain73.05 ± 21.6674.32 ± 20.590.5865.89 ± 24.8467.89 ± 21.980.500.170.33
 General health status63.32 ± 22.3873.17 ± 18.97< 0.00163.47 ± 22.7662.37 ± 26.080.970.960.14
 Vitality67.56 ± 17.5476.22 ± 14.610.00275.53 ± 16.5766.05 ±
 Social aspects79.44 ± 20.5688.56 ± 15.280.0183.74 ± 20.8381.74 ± 19.580.650.440.23
 Emotional aspects73.49 ± 33.5879.66 ± 31.580.5068.37 ± 40.8580.74 ±
 Mental health68.00 ± 18.9777.85 ± 16.450.0171.37 ± 24.1972.63 ± 19.60.490.330.36

Three patients had transitory joint pain for some time during the study; however, this did not require them to withdraw from the training program. Nine training group participants dropped out for personal reasons not related to health problems.

All EKG results were normal and the patients had no difficulty during the incremental protocol, with none of them stopping before the velocity of 7.0 km/hour. Thirty-nine percent of the training group participants and only 5.3% of the controls (P = 0.007) showed >15% improvement in VO2max. Forty-eight percent of training group participants and 5.3% of controls (P < 0.0001) improved at least 15% in anaerobic threshold VO2.


The present study demonstrated that a supervised exercise program improves aerobic capacity, functional capacity, exercise tolerance, oxygen pulse, fatigue, depression, and quality of life in patients with SLE. Patients who participated in the exercise program showed a significant improvement in functional capacity measured by the HAQ when compared with pretraining values, and also when compared with untrained controls. In contrast to the study by Tench et al (12), we did not detect a significant difference in the visual analog scale pain scores at the end of the study, probably because these scores were low at baseline.

Depression has been reported to occur in 31–52% of patients with SLE during some phase of life (37), and Krupp et al (4) detected a significant correlation between fatigue and depression. In the present study, a significant improvement in depression was observed after the exercise program, which is in contrast to a previous study that did not find a significant difference (20).

The low aerobic capacity occurring in SLE has been attributed to the low oxygen pulse during exercise (11), the lack of conditioning of the peripheral musculature (10), and fatigue (1, 12). Factors such as selective type II fiber atrophy (38–40) and lesions of the microcirculation (41) also can reduce oxygen extraction by muscle cells and interfere with the ability of oxygen extraction and utilization by the peripheral musculature during exercise.

There are only 3 studies on physical conditioning of patients with SLE that are available in the literature. In a pilot study, Robb-Nicholson et al (1) assessed physical conditioning and fatigue in 23 patients with SLE using a bicycle ergometer with concomitant gas analysis. Training was conducted in the patient's residence on a stationary bicycle for 8 weeks, with weekly supervision by telephone. No significant improvement in aerobic capacity was detected, although an improvement in fatigue was observed. Daltroy et al (19) studied 34 patients with SLE in a 12-week exercise program similar to that of Robb-Nicholson et al and obtained similar results with respect to improvement of fatigue, but they did not quantify the improvement in aerobic capacity.

Tench et al (20) evaluated 93 women with SLE divided into 3 groups. One group was instructed to perform exercise at home similar to that of the 2 previous studies, with supervision at 2-week intervals. The second group received relaxing therapy, and the third received no intervention. The authors used different speeds during an ergospirometric test on a treadmill, and observed improvement in fatigue and pain but not in aerobic capacity, depression, or quality of life in the group instructed to perform exercises. Due to lack of direct supervision for the physical conditioning programs in the previous studies, it is impossible to determine if the patients actually carried out the exercise according the program. These studies did not assess the anaerobic threshold, which is considered to be a safe measure of cardiovascular exercise by representing an important parameter for the determination of training HR and by respecting biologic individuality.

Despite previous studies that have shown that patients with fibromyalgia have a low aerobic capacity (42, 43), studies evaluating aerobic capacity in patients with SLE (6, 10–12, 44) and those assessing the effects of an exercise program (1, 19, 20), except for our previous study (45), did not exclude patients with a concomitant diagnosis of fibromyalgia, a fact that may have contributed to the low aerobic capacity detected in such patients.

In contrast to our results, the few studies available on exercise in SLE did not detect a significant improvement in aerobic capacity after intervention involving a program of aerobic exercise for patients with SLE (1, 19, 20). The use of a homogeneous protocol for all patients for the ergospirometric test on a computerized treadmill, the measurement of anaerobic threshold with the determination of individual training HR, the exclusion of patients with a concomitant diagnosis of fibromyalgia, and, mainly, the direct training supervision might have contributed to the improvement in the results.

The present study has limitations. The exclusion of patients with highly active disease or with severe organic impairment limits the generalization of the present results. Also, direct supervision requires qualified and experienced professionals. The training group patients maintained contact with the physiotherapist 3 times a week during the training program; however, this did not occur for the control group. We can not exclude the possibility that this additional attention could have been a factor in increased mood, motivation, and effort, which can influence the results.

In conclusion, supervised cardiovascular training is well tolerated and significantly improves exercise tolerance, aerobic capacity, oxygen pulse, fatigue, quality of life, depression, and functional capacity in patients with SLE. The importance of supervised aerobic exercises for clinically stable patients is therefore duly supported, suggesting that such a program can be offered at institutions that care for patients with SLE.


We thank Miss Yara Q. Confessor for collaboration during the ergospirometric tests in the exercise physiology laboratory. We also thank Dr. Branca D. B. Souza, from Santa Casa de Misericórdia de São Paulo, and Dr. William Chahade, from Hospital do Servidor Publico Estadual de São Paulo, for making available the recruitment of patients in the institutions.