Article Text

Download PDFPDF

Fatigue is associated with cerebral white matter hyperintensities in patients with systemic lupus erythematosus
  1. E Harboe1,
  2. O J Greve2,
  3. M Beyer2,
  4. L G Gøransson1,3,
  5. A B Tjensvoll4,
  6. S Maroni5,
  7. R Omdal1,3
  1. 1
    Clinical Immunology Unit, Department of Internal Medicine, Stavanger University Hospital, Stavanger, Norway
  2. 2
    Clinical Immunology Unit, Department of Radiology, Stavanger University Hospital, Stavanger, Norway
  3. 3
    Institute of Internal Medicine, University of Bergen, Bergen, Norway
  4. 4
    Clinical Immunology Unit, Department of Neurology, Stavanger University Hospital, Stavanger, Norway
  5. 5
    Clinical Immunology Unit, Department of Psychiatry, Stavanger University Hospital, Stavanger, Norway
  1. Erna Harboe, Stavanger University Hospital, Department of Internal Medicine, POB 8100, N-4068 Stavanger, Norway; hare{at}sus.no

Abstract

Background: Fatigue is a disabling phenomenon in many patients who have systemic lupus erythematosus (SLE). The pathophysiological processes are unknown, and no known biological disease factors influence the phenomenon. Because depressive mood is consistently associated with fatigue, and drug treatment for SLE does not ameliorate fatigue, a psychological explanation could be an alternative. In search of a somatic basis for fatigue, we looked for alternative markers of biologic activity associated with fatigue. Cerebral white matter hyperintensities (WMHs) represent biochemical changes of brain tissue and are frequently encountered in patients with SLE, and are associated with cognitive impairment in patients with multiple sclerosis. Presence of such an association between fatigue and WMHs in SLE would favour a biological axis to fatigue.

Methods: A cross-sectional, case–control study with 62 unselected patients with SLE and 62 age- and gender-matched healthy subjects. Fatigue was evaluated using the Fatigue Severity Scale (FSS) and a fatigue visual analogue scale (VAS). WMHs were rated using Scheltens’ method.

Results: Greater fatigue and more WMHs appeared in patients with SLE versus healthy subjects. In the full group of patients (n = 62), fatigue VAS was associated with total WMH score (p = 0.009). In subgroup analysis of patients without clinical depression (n = 40), the association with total WMH remained (p = 0.035), whereas this was not the case in the depressed group (n = 18) (p = 0.211).

Conclusion: Increased cerebral WMH load is associated with increased fatigue, indicating a biological origin for some portion of fatigue in patients with SLE.

Statistics from Altmetric.com

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease that may affect virtually any organ. Fatigue is prevalent and sometimes disabling in SLE, but it is not an established component of neuropsychiatric SLE syndromes (NPSLEs), as described by the American College of Rheumatology (ACR).1 Across a range of conditions, depressive mood is consistently reported to strongly influence fatigue, but it is uncertain whether there is a biological basis for fatigue.2 3

Cerebral white matter hyperintensities (WMHs), demonstrated on a T2-weighted magnetic resonance imaging (MRI) series, are more prevalent in patients with SLE. WMHs are not unique to SLE but occur in other diseases and in healthy people, with increased frequency occurring with increasing age and the presence of cerebrovascular risk factors.4 Some studies of patients with multiple sclerosis report an association between the number and size of WMHs—“the lesion load”—and cognitive dysfunction.5

The present study sought to characterise the extent and pattern of cerebral WMHs in an unselected SLE outpatient group compared with age- and gender-matched healthy subjects, and to determine if cerebral WMH load is associated with fatigue in patients with SLE.

METHODS

SLE patients

Medical records of all in- and outpatients diagnosed with SLE between 1980 and 2004 at the Stavanger University Hospital, Norway, were reviewed. Eighty-six patients, all Caucasians, fulfilled the revised ACR criteria for SLE.6 Sixty-two of these 86 patients (72%) were included: 54 women (87%) and 8 men (13%). The mean age (± SD) was 44.3 years (SD 13.2) (range 23.9–75.9 years), with a mean disease duration of 12.7 years (SD 8.8) (range 0.5–32.0 years). Fifty-two out of the 62 patients (84%) were not taking medication for SLE.

Nine patients (15%) had a clinical history of venous thromboembolic disease, and eight (13%) of cerebral stroke. Seven (11%) were well regulated on thyroxine-Na, and three (5%) had suffered a myocardial infarction. Three patients (5%) had a biopsy-confirmed, active glomerulonephritis, and one a renal transplant. One patient had moderate renal failure with a serum creatinine level of 187 μmol/L (reference interval 60–125 μmol/L).

Healthy subjects

Sixty-two age- (±2 years) and gender-matched individuals without neurological, immunological or malignant disease served as healthy control subjects. The volunteers were recruited from friends and members of the hospital staff or from non-familial friends of the patients.

Clinical assessment and evaluation of fatigue and depression

All participants were subjected to a standardised general medical and neurological examination. Disease activity was measured with the SLE Disease Activity Index (SLEDAI),7 and organ damage was measured with the Systemic Lupus International Collaborating Clinics/ACR damage index (SLICC).8 Fatigue was assessed with the Fatigue Severity Scale (FSS) and a fatigue visual analogue scale (VAS).3 The Beck Depression Inventory (BDI) was used to assess mood. A cut-off score ⩾13 was used to detect clinical depression.9

Cerebral MRI

Patients were examined in a 1.5-T Philips Gyroscan NT Intera Release 10. For this study, the axial T2 and sagittal FLAIR were used. The interval between clinical and MRI examinations was 12 days (SD 11) for SLE patients and 28 days (SD 24) for the healthy subjects.

Four regions of cerebral white matter were evaluated using the semi-quantitative scale of Scheltens’: the periventricular region (PV); the subcortical and deep white matter of the frontal, parietal, temporal and occipital areas (FPTOs); the basal ganglia (BG); and the infratentorial region (IT).10 This method takes into account the spatial localisation, size and number of lesions. The total WMH sum score range 0–84 was previously used to measure WMH burden.11

All scans were rated by two experienced radiologists (MB and OJG) in a blinded manner; the inter- and intrarater reliabilities for WMH ratings were good to excellent.12

Statistical methods

Results are reported as mean ±SD when normally distributed, otherwise as mean ±SD with median and range. Pairwise comparison was performed using paired t-test (two-tailed) for continuous and McNemar’s test for categorical data. Simple or multiple regression analysis was used to assess associations between one dependent and one or several independent continuous variables. A significance level of 0.05 was used, and p values were corrected for ties. Bonferroni correction was applied for multiple comparisons, when appropriate.

Stepwise linear regression with backward selection was used to evaluate the influence of independent variables on FSS and fatigue VAS.

RESULTS

The disease activity score (SLEDAI) was 3.5 (SD 2.8) (median 2.0, range 0–12.0), which is compatible with a low to medium disease activity. The damage index score (SLICC) was 2.4 (SD 2.2) (median 2.0, range 0–11.0).

FSS and fatigue VAS scores were higher in patients than in healthy subjects (FSS: 4.3 (SD 1.7) (median 4.6, range 1.3–7.0) vs 2.3 (SD 1.0) (median 2.1, range 1.0–5.3), p<0.0001; fatigue VAS: 48.9 (SD 30.1) (median 50, range 1–98) vs 19.7 (SD 19.7) (median 13, range 1–72), p<0.0001). The correlation between FSS and fatigue VAS was 0.79 (p<0.0001). There were more depressive symptoms in patients than in healthy subjects (BDI 8.4 (SD 7.2) (median 8, range 0–27) vs 2.5 (SD 2.6) (median 2, range 0–9), p<0.0001). Eighteen patients (31%), but no healthy subjects, had clinically relevant depression (BDI score ⩾13; p<0.0001). The haemoglobin concentration (g/dL) and thyroid-stimulating hormone (TSH; mIU/L) differed between patients with SLE and healthy subjects (13.1 (SD 1.1) vs 13.7 (SD 0.9), p = 0.001; 2.9 (SD 2.7) vs 2.1 (SD 0.9), p = 0.038). Hypertension (systolic blood pressure >140 mm Hg, diastolic blood pressure >90 mm Hg, or receipt of antihypertensive treatment) was more prevalent among patients with SLE than healthy subjects (31 vs 19, p = 0.008).

Cerebral MRI

Assessment by MRI revealed one or more cerebral infarcts in eleven patients (18%) and two healthy subjects (p = 0.012). Total WMH scores increased more with age in patients with SLE than in healthy subjects (fig 1).

Figure 1 Bivariate scatterplot with total white matter hyperintensities (WMHs) in relation to age in 62 patients with systemic lupus erythematosus (SLE) and 62 healthy subjects matched for age and gender. The lines represent WMHs in patients with SLE and healthy subjects according to simple regression analysis in the two groups. The equations for the regression lines were as follows: Total WMH in SLE patients  =  −7.8+0.31*Age, R2 = 0.37, p<0.0001. Total WMH in healthy subjects  =  −6.7+0.24*Age, R2 = 0.33, p<0.0001.

Total WMH scores were higher in patients than in healthy subjects (6.0 (SD 6.8) vs. 4.1 (SD 5.7), p = 0.050). WMH scores for the FPTO region were higher in patients with SLE (patients 3.6 (SD 4.6) vs. healthy subjects 2.0 (SD 3.0); p = 0.042). There was no group difference for the PV region, and there were too few observations to perform group comparisons for the BG and IT regions.

Associations between fatigue and clinical variables

There were no associations between fatigue and the use of medication for SLE, including the current dose of prednisolone. Other variables are presented in table 1.

Table 1 Associations between fatigue and selected independent variables in simple regression analysis with known* or suspected impact on fatigue

Associations between fatigue and MRI findings

Fatigue VAS, but not FSS score, was associated with total WMH score (y = 40.2+1.5x, R2 = 0.11, p = 0.009). Both fatigue VAS (y = 33.5+9.5x, R2 = 0.24, p<0.0001) and FSS (3.79+0.33x, R2 = 0.09, p = 0.021) were associated with PV score. No such association was evident in the FPTO region.

In subgroup analysis of SLE patients without clinical depression, BDI scores <13 (n = 40), the association between fatigue VAS and total WMH (y = 29.6+1.3x, R2 = 0.11, p = 0.035) and PV region (y = 26.3+ 8.1x, R2 = 0.19, p = 0.005) was significant, whereas this was not the case for the depressed patients (n = 18) (p = 0.211 and p = 0.138). There was no association between fatigue VAS and FPTO region in subgroup analysis of non-depressed (p = 0.086) or depressed (p = 0.412) patients. FSS was not associated with total WMH, FPTO or PV region in subgroup analysis.

Impact of multiple variables on fatigue

Stepwise regression analysis was performed with the independent variables BDI, haemoglobin, total WMH, TSH, age, SLEDAI and SLICC. In the total patient group who completed BDI (n = 58), consisting of both depressed and non-depressed individuals, haemoglobin, BDI and total WMH were significantly associated with fatigue VAS (final model after four steps: adjusted R2 = 0.44; p<0.0001) for FSS; only BDI stayed in the model (R2 = 0.38, p<0.0001).

DISCUSSION

This study confirms previous findings that the prevalence and magnitude of fatigue is considerably higher in patients with SLE than in healthy subjects.2 3 13 It provides further evidence that fatigue is not associated with disease activity variables found in the SLEDAI, but reveals an association between organ damage (SLICC) and fatigue. The latter finding contrasts two previous studies.13 14 Moreover, the study extends the understanding of fatigue by demonstrating that fatigue is associated with a higher cerebral WMH load in patients with SLE, supporting evidence for a biological basis for some portion of fatigue in patients with chronic autoimmune diseases.

Consistent with previous findings, we found depressive mood measured by BDI to be the strongest factor influencing fatigue.13 However, disabling fatigue was also evident among our patients who did not suffer from depression. This indicates that other and probably more disease-specific mechanisms are involved.

The finding of significantly higher WMH scores in SLE patients than in healthy subjects indicates a disease-associated pathogenesis for WMH and are in line with reports of elevated markers of ongoing damage to neurons and astrocytes in the cerebrospinal fluid of patients with NPSLE.15 A similar association between WMH and fatigue in patients with multiple sclerosis complements our findings.16

Acknowledgments

The authors would like to thank Jan Terje Kvaløy, PhD, Department of Mathematics and Natural Sciences, University of Stavanger, for statistical advice.

REFERENCES

Footnotes

  • Funding: Lasse G. Gøransson received support as a doctoral research fellow from Western Norway Regional Health Authority.

  • Competing interests: None declared.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.