Obesity induces a proinflammatory state and is a major cause of morbidity in the general population. However, little is known about the effects of obesity in patients with chronic inflammatory illnesses such as systemic lupus erythematosus (SLE).
One hundred consecutive patients with SLE were studied to determine the relationship between body mass index (BMI) and functional capacity, measures of fatigue, quality of life, and the inflammation markers C-reactive protein (CRP), the erythrocyte sedimentation rate, and interleukin-6 (IL-6). The association between BMI and patient characteristics was determined, and multiple logistic regression models were used to adjust for age, sex, disease activity, and disease-related damage.
Thirty-three patients had a normal BMI (<25 kg/m2), 28 were overweight (25–29.9 kg/m2), and 39 were obese (≥30 kg/m2). Obese patients had worse functional capacity, more fatigue, and higher concentrations of inflammation markers. The mean ± SD modified Health Assessment Questionnaire (M-HAQ) score was 0.6 ± 0.4 in obese patients compared with 0.3 ± 0.4 and 0.2 ± 0.3 in overweight patients and those with a normal BMI, respectively (P = 0.001). The mean ± SD concentrations of CRP in obese patients (10.0 ± 8.6 mg/liter) were higher than those in patients who were overweight (4.7 ± 5.4 mg/liter) or had a normal BMI (6.2 ± 9.9 mg/liter) (P < 0.001). Similarly, concentrations of IL-6 were higher in obese patients (P = 0.003). After adjusting for age, sex, disease activity, and damage indices, the associations between BMI and CRP (P < 0.001), M-HAQ scores (P = 0.005), and IL-6 concentrations (P = 0.01) remained significant.
Obesity is independently associated with impaired functional capacity and inflammation markers in patients with lupus. Thus, weight loss may improve functional capacity and decrease cardiovascular risk factors.
Two of every 3 adults in the US are overweight or obese (1), and the prevalence is increasing (2). Obesity is a major risk factor for increased mortality in the general population (3); it is estimated that in the US alone, 300,000 premature deaths annually are related to obesity (4). Furthermore, obesity increases the cost of health care (5), resulting in at least $90 billion in direct healthcare costs in the US (4). Interest in the adverse consequences of obesity has focused on increased cardiovascular risk (and the role of inflammation mediators in this risk) and also on diabetes and impaired quality of life.
Several studies have shown that obesity contributes independently to cardiovascular risk. For example, prospective data from the Framingham Heart Study showed that weight gain increased cardiovascular risk, and that obesity was a risk factor for coronary heart disease (6). The risk of coronary heart disease was increased >2-fold in men with a body mass index (BMI) of 29.0–32.9 kg/m2 and >3-fold in those with a BMI of ≥33 kg/m2 (7). Obesity also increased mortality due to cardiovascular disease, with relative risks of 1.46 and 2.07, respectively, in men and women in the highest quartiles for BMI (8). The risk of coronary heart disease in the general population is associated with increased concentrations of markers and mediators of inflammation such as C-reactive protein (CRP) and interleukin-6 (IL-6); concentrations of both CRP and IL-6 are also elevated in the setting of obesity (9, 10). Thus, there is speculation that the increased inflammation associated with obesity may contribute to the associated increase in cardiovascular risk.
In addition to increasing cardiovascular risk, obesity increases the risk of diabetes and has an adverse effect on quality of life. The Nurses' Health Study showed that the risk of diabetes was lowest among individuals with a BMI <22 kg/m2 and increased as the BMI increased (11). Similarly, in a large prospective cohort of male professionals, those with a BMI of ≥35 kg/m2 had a multivariate relative risk of 42.1 compared with those with a BMI <23 kg/m2 (12). In addition to increasing the risk of serious medical problems, BMI is inversely associated with physical aspects of quality of life among different populations of varying ages, and obese individuals have decreased physical function and higher levels of pain (13–17).
There is little information about the effects of obesity on symptoms, functional capacity, and markers of inflammation in patients with systemic lupus erythematosus (SLE). This is important because patients with SLE have chronic inflammation, impaired quality of life, and increased cardiovascular risk, and it is not known whether obesity is a contributing factor. Therefore, we examined the hypothesis that obesity contributes to impaired functional capacity, symptoms that affect quality of life, and increased concentration of inflammation markers in patients with SLE.
PATIENTS AND METHODS
The study group comprised 100 consecutive eligible patients, age 18 years or older, who met the classification criteria for SLE (18) and had a disease duration of >1 year. Patients were recruited from the practices of local rheumatologists, through a Lupus Foundation newsletter, and by advertisements. The study was approved by the institutional review board of Vanderbilt University Hospital, and all patients provided written informed consent.
Patients were studied at the General Clinical Research Center at Vanderbilt University. Information was obtained using a structured interview, physical examination, laboratory tests, and review of medical records. BMI was calculated, and patients were divided into 3 groups as follows: BMI <25 kg/m2 (normal), BMI 25–29.9 kg/m2 (overweight), and BMI ≥30.0 kg/m2 (obese) (19). Physical activity was recorded as the number of exercise sessions per week and the duration of each session. Based on The American College of Sports Medicine definition of fitness (20), patients were classified into 1 of 3 exercise categories as follows: none (no exercise at all), some (<3 times per week and/or <30 minutes per session), and fit (≥3 times per week and ≥30 minutes per session).
Fasting blood samples were collected for a complete blood cell count and determination of creatinine, total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, Lp(a), homocysteine, and CRP levels, the CH50 concentration, and the Westergren erythrocyte sedimentation rate (ESR). Plasma was separated by centrifugation and stored at −70°C. Plasma IL-6 concentrations were determined by enzyme-linked immunosorbent assay using commercially available kits (R&D Systems, Minneapolis, MN), and results were reported as picograms per milliliter. For technical reasons, IL-6 concentrations were determined in only 81 patients.
Indices of disease activity and damage.
Two composite lupus-specific disease indices were used. Disease activity was ascertained using the SLE Disease Activity Index (SLEDAI), (21) and cumulative organ damage was determined by the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) (22).
Measures of functional capacity, attitudes, helplessness, pain, and fatigue.
Patients were asked to indicate on a 10-cm visual analog scale (VAS) the point that best described their fatigue and pain, with higher scores indicating worse symptoms. Patient questionnaire measures also included several commonly used instruments.
The modified Health Assessment Questionnaire (M-HAQ), a standard 8-question instrument, was used to assess functional capacity based on the difficulty in performing activities of daily living. This questionnaire is scored from 0 to 3, with higher scores indicating lower functional capacity (23).
The brief Rheumatology Attitudes Index (RAI) is a 5-item instrument used to assess the construct of helplessness, defined as a psychological state in which an individual expects that his or her efforts will be ineffective. It is scored from 1 to 5, and higher scores indicate higher levels of helplessness (24).
The Krupp Fatigue Severity Scale is a 9-item questionnaire that was previously validated as a measure of fatigue severity in patients with SLE. The severity of fatigue is rated from 1 to 7, with higher scores indicating more severe levels of fatigue (25).
The Spitzer Quality of Life Index is a concise instrument that evaluates a positive mood state, supportive relationships, and the absence of physical and psychological distress. Scores range from 0 to 10, with higher scores indicating better quality of life (26).
All of these self-report measures of outcome have been used in patients with SLE and patients with other rheumatic conditions (24–29).
Demographic characteristics are presented as the mean ± SD for continuous variables and as frequencies and percentages for categorical variables. Patients were classified into 3 BMI categories, as follows: obese (≥30.0 kg/m2), overweight (25.0–29.9 kg/m2), and normal (<25 kg/m2) (19). Univariate analyses were performed to compare differences among patients in the 3 BMI groups. Overall differences within the groups were compared using Kruskal-Wallis tests for continuous variables and Pearson's chi-square tests for categorical variables. Spearman's correlation coefficients were used to determine the association between BMI and clinical characteristics.
Multiple logistic regression models were used to determine the association between BMI and patient outcome measures and markers of inflammation, by comparing the BMI in patients whose values for each outcome variable were above and below the median. To test whether these associations were independent of age and sex, we adjusted for these factors and then further adjusted for disease activity and disease damage. All analyses used a 2-sided significance level of 5% and were performed using Stata version 8.2 software (College Station, TX).
One hundred patients with SLE (90 women and 10 men) were studied. Their mean ± SD age was 40.7 ± 12.3 years, the mean ± SD duration of disease 9.4 ± 8.6 years, and the mean ± SD BMI of the patients was 28.6 ± 6.4 kg/m2. Seventy-one patients were white, 26 were African American, and 3 were of other ethnicity. Ninety-six patients had completed high school. Thirty-three patients had a normal BMI (<25 kg/m2), 28 were overweight (25–29.9 kg/m2), and 39 were obese (≥30 kg/m2).
Table 1 compares demographic and disease activity variables and inflammation markers in patients categorized according to BMI. Obese patients were older, with a mean ± SD age of 43.6 ± 12.3 years compared with 41.3 ± 12.5 years in the overweight group and 36.9 ± 11.4 years in patients with a normal BMI. There were significant differences in disease duration between the 3 groups (9.2 ± 8.8 years for patients with a normal weight, 12.4 ± 9.0 years for overweight patients, and 7.3 ± 7.5 years for obese patients [P = 0.007]). The groups did not differ significantly with regard to sex, race, or level of education. Hemoglobin and creatinine concentrations were similar in the 3 groups, but blood glucose concentrations were higher in overweight and obese patients. The prevalence of hypertension was increased in overweight and obese patients compared with that in patients with a normal BMI. There were no significant differences among the 3 groups of patients in disease activity (P = 0.40) or cumulative damage (P = 0.99) as determined by SLEDAI and SDI scores, respectively. However, concentrations of CH50 were higher among obese patients. There was no significant difference between the groups in the proportion of patients receiving corticosteroids or cumulative exposure to these drugs. The frequency of each category of exercise did not differ significantly between patients in the 3 categories of BMI (P = 0.16). One patient in the obese group and 1 patient in the overweight group had a history of myocardial infarction or angina (P = 0.74).
Table 1. Characteristics of patients with SLE according to BMI category*
Normal (n = 33)
Overweight (n = 28)
Obese (n = 39)
Except where indicated otherwise, values are the mean ± SD. Kruskal-Wallis tests were used for continuous variables, and Pearson's chi-square tests were used for categorical variables. See Patients and Methods for a description/explanation of patient questionnaire measures. Values were missing as follows: for metabolic syndrome, n = 9; for C-reactive protein, n = 2; for CH50, n = 4; for interleukin-6, n = 19; for total cholesterol, n = 2; for high-density lipoprotein, low-density lipoprotein, triglycerides, and Lp(a), n = 2. SLE = systemic lupus erythematosus; BMI = body mass index; SLEDAI = SLE Disease Activity Index; SLICC/ACR = Systemic Lupus International Collaborating Clinics/American College of Rheumatology; M-HAQ = modified Health Assessment Questionnaire; VAS = visual analog scale.
As determined by a single question on the Multidimensional HAQ.
CRP and IL-6 concentrations and the ESR were increased in obese patients (Table 1). Obese patients had significantly decreased functional capacity as reflected by higher M-HAQ scores, increased scores for pain and fatigue, and decreased scores for quality of life and helplessness. There were no significant differences between white and nonwhite patients for the M-HAQ score (P = 0.71), the pain score on a VAS (P = 0.12), the disease activity score on a VAS (P = 0.96), the Quality of Life Index (P = 0.53), the RAI (P = 0.90), the Fatigue Severity Scale (P = 0.22), the IL-6 concentration (P = 0.73), and the CRP level (P = 0.20).
Table 2 shows the relationship between the BMI and clinical characteristics of the patients. BMI was significantly correlated with age (ρ = 0.26, P = 0.01), systolic blood pressure (ρ = 0.30, P = 0.003), HDL cholesterol (ρ = −0.24, P = 0.02), and triglycerides (ρ = 0.27, P = 0.007). BMI also correlated significantly with markers of inflammation such as the ESR (ρ = 0.21, P = 0.03), the CRP level (ρ = 0.43, P < 0.001), and the IL-6 concentration (ρ = 0.33, P = 0.003) and was associated with poorer physical function, as determined by the M-HAQ (ρ = 0.34, P < 0.001), higher levels of fatigue (ρ = 0.31, P = 0.002), poorer attitudes (ρ = 0.27, P = 0.007), poorer quality of life (ρ = −0.26, P = 0.009), and higher levels of pain (ρ = 0.24, P = 0.02). M-HAQ scores and CRP and IL-6 concentrations in the 3 groups are shown in Figure 1.
Table 2. Correlation between body mass index and clinical characteristics*
SLICC/ACR = Systemic Lupus International Collaborating Clinics/American College of Rheumatology; VAS = visual analog scale.
Disease duration, years
Systolic blood pressure, mm Hg
Diastolic blood pressure, mm Hg
Total cholesterol, mg/dl
High-density lipoprotein cholesterol, mg/dl
Low-density lipoprotein cholesterol, mg/dl
Cumulative corticosteroid exposure, gm
Erythrocyte sedimentation rate, mm/hour
C-reactive protein, mg/liter
Systemic Lupus Erythematosus Disease Activity Index
SLICC/ACR Damage Index
Patient questionnaire measures
Modified Health Assessment Questionnaire (0–3 scale)
Pain score (0–10-cm VAS)
Fatigue score (0–10-cm VAS)
Quality of Life Index (10–0 scale)
Rheumatology Attitudes Index
Fatigue Severity Scale (1–7 scale)
Table 3 shows the BMI in patients, stratified according to values above and below the median for the variables of interest. An increased BMI was associated with a higher risk of being in the group with worse functional capacity and poor quality of life, higher levels of pain, fatigue, and helplessness, and higher concentrations of CRP and IL-6. After adjusting for age, sex, disease activity, and damage, BMI remained associated with higher M-HAQ scores (odds ratio [OR] 1.12, P = 0.005) and higher IL-6 (OR 1.11, P = 0.01) and CRP concentrations (OR 1.20, P < 0.001). The significant independent associations between obesity and M-HAQ, CRP, and IL-6 did not change after including race, menopause status, or corticosteroid use in the models (P < 0.05 for all).
Table 3. Body mass index (BMI) in patients, stratified according to values above and below the median for each variable*
Unadjusted OR (95% CI)
OR adjusted for age and sex (95% CI)
OR adjusted for age, sex, SLEDAI, and SDI
Except where indicated otherwise, values are the mean ± SD. OR = odds ratio; SLEDAI = Systemic Lupus Erythematosus Disease Activity Index; 95% CI = 95% confidence interval; SDI = Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; HAQ = Health Assessment Questionnaire; VAS = visual analog scale; RAI = Rheumatology Attitudes Index.
Modified HAQ score
≤0.25 (n = 56)
26.4 ± 5.1
>0.25 (n = 44)
31.3 ± 6.9
Pain score on VAS
≤4 (n = 50)
27.2 ± 5.6
>4 (n = 50)
30.0 ± 6.9
Fatigue score on VAS
≤5 (n = 48)
26.9 ± 5.0
>5 (n = 51)
30.1 ± 7.2
Fatigue Severity Scale
≤5 (n = 46)
27.0 ± 4.9
>5 (n = 53)
29.9 ± 7.3
Quality of life Index
>8 (n = 42)
26.8 ± 5.9
≤8 (n = 57)
29.8 ± 6.5
≤2.5 (n = 57)
27.7 ± 5.8
>2.5 (n = 42)
29.7 ± 7.1
C-reactive protein, mg/liter
<4 (n = 42)
25.3 ± 4.9
≥4 (n = 5)
31.0 ± 6.4
≤8 (n = 40)
26.2 ± 4.3
>8 (n = 41)
30.4 ± 7.6
The major new findings of this study are that two-thirds of patients with SLE were either overweight or obese, and that obese patients were more likely to experience poorer functional capacity, impaired health-related quality of life, higher levels of fatigue, and worse pain. In addition, patients who were overweight or obese had higher concentrations of markers of inflammation.
The mechanisms underlying increased inflammation in the setting of obesity are not clear, but mononuclear cells are activated in obese individuals and produce proinflammatory cytokines (30). Also, adipocytes produce cytokines, complement components, and other proinflammatory mediators (31). Among patients with SLE, as in the general population, CRP concentrations were elevated in obese individuals. This finding was unexpected, because in the Women's Health Study the median CRP concentration in obese women was ∼3 mg/liter (32); this concentration is much lower than that in patients with SLE, a chronic inflammatory disease, in whom the mean ± SD CRP concentration was 7.3 ± 8.6 mg/liter. Thus, it seemed likely that the relatively small contribution of obesity to increased CRP concentrations would be obscured by the larger contribution of SLE. This was not the case; CRP concentrations were higher in obese patients with lupus, and, as expected, there was a moderate correlation between CRP and BMI; this finding suggests that obesity may contribute significantly to the overall inflammation burden in patients with SLE. This observation is important, because recent findings suggest that cardiovascular risk increases linearly with the CRP level, even at very high concentrations (33). Furthermore, elevated concentrations of CRP were an independent predictor of vascular events in patients with lupus (34). Thus, our finding that obese patients with lupus have higher concentrations of both CRP and IL-6 raises concern that these patients may be at particular risk of vascular disease.
The observation that the CH50 concentration was higher in obese patients with SLE than in patients who were overweight or had a normal BMI is also new. Concentrations of complement increase in response to inflammation, and information from the general population showed that concentrations of complement C3 are associated with obesity (35), and that C3 is produced in part by adipocytes and is associated with insulin resistance and coronary disease (36, 37). Thus, our findings showing a relationship between obesity and higher concentrations of complement in patients with SLE, a population in whom low complement concentrations may impart risk or indicate active lupus, are concordant with findings in the general population.
The relationship between obesity and elevated concentrations of blood glucose is well recognized (11, 12). Even though mean blood glucose concentrations were normal in all 3 groups of patients with SLE, concentrations of glucose were significantly higher in obese patients. Elevated concentrations of glucose within the normal range are clinically important, because higher concentrations of fasting glucose are associated with insulin resistance, a major component of the metabolic syndrome that is linked to cardiovascular disease (38, 39).
Numerous adverse events, including central obesity and glucose intolerance, have been reported in patients receiving corticosteroids, especially those receiving high doses for a long period of time. We therefore examined the relationship between corticosteroid exposure and BMI and found no relationship between current or cumulative exposure to corticosteroids and BMI. Thus, exposure to corticosteroids is not a major determinant of BMI in patients with SLE.
In addition to the effects of increased BMI on mediators of inflammation and blood glucose concentrations, increased BMI also has profound effects on quality of life. In the general population, an increased BMI was associated with impaired physical function, pain, low vitality, and poor quality of life (13, 15, 16, 40). However, little is known about the effects of obesity on functional and psychological measures of quality of life in patients with SLE. Several studies have shown that self-reported functional outcome and quality of life measures are significantly lower in patients with lupus than in the general population. The majority of studies that used disease activity instruments that are independent of quality of life measures showed that disease activity and damage contribute relatively little to these outcomes (41–47). Our findings indicate that obesity is an important independent risk factor associated with poor outcomes in patients with lupus.
Even modest obesity can have a significant impact on quality of life, particularly in women (40). Generally, obesity affects the physical components of quality of life more than the psychological components (48). However, an effect of proinflammatory cytokines on emotional behavior has been suggested in both animal models and human studies (49–51). Thus, the fact that obese patients have increased concentrations of proinflammatory cytokines may contribute to impairment of the psychological quality of life measures. A possible mechanism to explain the contribution of obesity to inflammation and quality of life is the association of obesity with the metabolic syndrome. As expected, based on the National Cholesterol Education Program definition of the metabolic syndrome (52), our results suggest that the metabolic syndrome was present in 24% of overweight patients and in 52.8% of obese patients, compared with only 10% of patients whose weight was normal. Further studies, including assessment of insulin sensitivity, would be of interest to investigate the association between the metabolic syndrome, inflammation, and other cardiovascular risk factors in patients with SLE.
Results of a previous study suggested a trend toward an association between the number of tender points and functional capacity (53). Thus, the coexistence of fibromyalgia may be a potential confounding variable. However, the study protocol did not include the assessment of tender points, and thus we cannot define the contribution of fibromyalgia to our observations.
We adjusted for several variables that could affect concentrations of inflammation mediators and quality of life. For example, age can affect both quality of life and concentrations of inflammation mediators. In this study, obese patients (average age 43.6 years) were older than those with a normal BMI (average age 36.9 years), but these differences would not be sufficient to explain our results, because HAQ scores in the general population do not change much before age 45 years (54), and concentrations of inflammation markers such as CRP are minimally affected by differences of up to 10 years in age (55). Furthermore, even after adjusting for age and sex, the association between BMI and the M-HAQ scores, fatigue scores, and concentrations of inflammation markers remained significant.
The identification of obesity as a cause of impaired quality of life and a contributor to inflammation in SLE is important, because obesity is a risk factor that can be modified, and such modification alters risk (56). In the general population, reducing obesity has improved function and decreased pain and concentrations of markers of vascular inflammation. For example, exercise, dietary modification (14), and weight loss (57) improved health-related quality of life, and lifestyle changes designed to reduce body weight in obese women decreased concentrations of IL-6 and CRP (58).
This study has several limitations. Even when the association is significant, the temporal relationship between obesity, markers of inflammation, and functional capacity is unknown, and thus a causal relationship cannot be proven. The relationship between obesity and impaired functional capacity is difficult to interpret using cross-sectional data. Obesity could be either a cause or a result of impaired functional capacity. However, studies in the healthy population have shown that weight loss in obese persons improved functional capacity and vascular markers of inflammation, suggesting that it is more likely that obesity results in impaired functional capacity and inflammation (58, 59). Similarly, increased concentrations of inflammation markers such as IL-6 and CRP are associated with both obesity (60, 61) and SLE (62, 63). The mean concentration of CRP in lupus patients of normal weight (6.2 mg/liter) is substantially higher than that in the healthy population (1.35 mg/liter) (64). In apparently healthy obese women (average BMI 33 kg/m2) the mean CRP concentration was ∼7.0 mg/liter (65); this value is lower than that in obese patients with lupus (10.0 mg/liter). These findings suggest that both lupus and obesity contribute to the inflammation burden in patients with SLE.
Our study design did not provide measurements over time; persistent levels of inflammation and decreased functional capacity in obese patients would strengthen the possibility of a causal relationship. Correlations should be interpreted cautiously, particularly when rho values are between 0.20 and 0.30 (relatively weak). In addition, the number of patients studied may be insufficient to demonstrate other significant associations.
In conclusion, we found that obesity is common in patients with SLE and is independently associated with impaired functional capacity and increased inflammation. These findings suggest that interventions to reduce or reverse obesity could improve both symptoms and long-term outcomes of patients with lupus.
We are indebted to Carol Brannon, who assisted with recruitment of subjects and helped enter data, and to Dr. Theodore Pincus for a thoughtful review of the manuscript.