Significantly Higher Percentage of Circulating CD27high Plasma Cells in Systemic Lupus Erythematosus Patients with Infection than with Disease Flare-Up

Deng-Ho Yang; Deh-Ming Chang; Jenn-Haung Lai; Fu-Huang Lin; Chen-Hung Chen

doi:10.3349/ymj.2010.51.6.924

Journal List > Yonsei Med J > v.51(6) > 1030727

Go to TopGo to Top Go to BottomGo to Bottom

TOOLS

Yang, Chang, Lai, Lin, and Chen: Significantly Higher Percentage of Circulating CD27high Plasma Cells in Systemic Lupus Erythematosus Patients with Infection than with Disease Flare-Up

Original Article

Rheumatology

Yonsei Medical Journal

Yonsei Medical Journal 2010; 51(6): 924-931.

Published online: 30 September 2010

DOI: https://doi.org/10.3349/ymj.2010.51.6.924

Significantly Higher Percentage of Circulating CD27^high Plasma Cells in Systemic Lupus Erythematosus Patients with Infection than with Disease Flare-Up

Deng-Ho Yang^1,², Deh-Ming Chang², Jenn-Haung Lai², Fu-Huang Lin³, Chen-Hung Chen²

¹Division of Rheumatology/Immunology/Allergy, Department of Internal Medicine, Armed-Forces Taichung General Hospital, Taichung, Taiwan, Republic of China.

²Division of Rheumatology/Immunology/Allergy, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, Republic of China.

³Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan, Republic of China.

Corresponding author: Dr. Chen-Hung Chen, Division of Rheumatology/Immunology/Allergy, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, No. 325, Sec. 2, Cheng-Kung Rd, Neihu 114, Taipei, Taiwan, Republic of China. Tel: 886-2-87927135, Fax: 886-2-87927136, deng6263@ms71.hinet.net

Received 26 August 2009 Revised 1 February 2010 Accepted 9 February 2010

(open-access):

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Purpose

To distinguish lupus flare-up from infection in systemic lupus erythematosus (SLE), we analyze the expression of circulating CD27^high plasma cells in SLE patients with and without infection, in comparison to non-SLE patients with infection.

Materials and Methods

The percentage of circulating CD27^high plasma cells was measured by flow cytometry in the following four groups: 36 SLE patients without infection, 23 SLE patients with infection, eight non-SLE patients with infection, and 26 healthy controls.

Results

The frequency of CD27^high plasma cells had a correlation with the SLE disease activity index (SLEDAI) (r = 0.866, p < 0.05), level of anti-dsDNA (r = 0.886, p < 0.05), C3 (r = - 0.392, p < 0.05), and C4 (r = - 0.337, p < 0.05) in SLE patients without infection, but there was no correlation with disease activity in SLE patients with infection. Among three groups in particular-SLE without infection, SLE with infection, and non-SLE with infection-the percentages of CD27^high plasma cells were elevated. The percentage of CD27^high plasma cells was higher in SLE patients with infection, when compared to SLE patients without infection.

Conclusion

The percentage of CD27^high plasma cells is a biomarker for disease activity of SLE without infection, under correlation with SLEDAI, anti-dsDNA, and C3 and C4 level. However, when the SLE patients have an infection, the percentage of CD27^high plasma cells is not an adequate biomarker for the survey of disease activity. The percentage of CD27^high plasma cells may serve as a potential parameter to distinguish a lupus flare-up from infection.

Keywords: Systemic lupus erythematosus, infection, CD27

INTRODUCTION

Systemic lupus erythematosus (SLE) is an autoimmune disorder with multiorgan involvement; it is characterized by the production of numerous autoantibodies.1,2 Although the pathogenesis of SLE remains unclear, autoantibodies against autoantigen, the deposition of immune complexes, and activation of the classical pathway complement system seem to be consequences of immune dysregulation.2,3 Abnormal immune responses involve innate and adaptive immune systems. These mechanisms-including hyperactive memory B cells and plasma cells, abnormal T cell activity, monocyte abnormalities, and immunoregulation-play important roles in the development of SLE.1,2,4-6 In patients with SLE, hyperactive B cells are triggered by an autoantigen or external antigen, to produce numerous autoantibodies that appear to have pathogenic consequences.7 These autoantibodies can induce damage to targeted organs and structures, including the skin, lung, brain, blood vessels, heart, kidneys, and joints.

Previous studies describe many new biomarkers for the survey of SLE; these potential lupus biomarkers can be used to survey susceptibility, diagnosis, disease activity, and specific organ involvement.8 The novel biomarkers to evaluate disease activity include serum cytokine, soluble cytokine receptor, soluble cell surface molecules (CD27, CD154 and BAFF), endothelial activation markers (sVCAM, sICAM and thrombomodulin), and cell markers (CD27^high plasma cells and erythrocyte-C4d).8,9 Traditional biomarkers for the survey of disease activity include anti-dsDNA antibodies and serum complement levels (C3 and C4). However, a persistently high level of anti-dsDNA antibodies or a low level of complement (C3 and C4) is found in some patients with SLE. Therefore, traditional biomarkers are not always adequate in evaluating disease activity. A potential biomarker is required to survey disease activity.

Previous studies have proposed that the frequency of CD27^high plasma cells is high in SLE patients, and that this metric is useful in evaluating disease activity with a significant correlation with the systemic lupus erythematosus disease activity index (SLEDAI) and autoantibodies.10,11 However, the percentage of circulating CD27^high plasma cells is also increased in non-SLE patients with bacterial infection.12 Distinguishing active lupus from infection is traditionally difficult. Elevated C-reactive protein (CRP) levels over 5 mg/dL suggest the presence of infection.13,14 Our study aims to evaluate the percentage of circulating CD27^high plasma cells from the peripheral blood in the SLE patient, either with or without infection.

MATERIALS AND METHODS

Study participants

All study participants were 18 years of age or above, and each provided written informed consent. None of the patients were excluded from participation on the basis of sex or ethnicity. This study was approved by the Tri-Service General Hospital Institutional Review Board (Grant no: 098-05-084) based on the declaration of Helsinki.

SLE patients

We analyzed 59 patients (52 women and 7 men) with SLE, all of whom fulfilled the 1982 revised criteria of the American College of Rheumatology, after informed consent was obtained. Disease activity was evaluated in each patient according to the SLEDAI. In laboratory studies, the following characteristics were investigated in the patients with SLE: CRP, blood urea nitrogen, creatinine, white blood cell count, haemoglobulin, platelet, anti-dsDNA, anticardiolipin IgG/IgM, anti-Sm, anti-Ro, anti-La, anti-RNP, rheumatoid factor, and C3 and C4. The flare-up of SLE was defined as the elevation over 3 points in SLEDAI (compared with the patient's previous SLEDAI) without evidence of infection.

Non-SLE patients with bacterial infections

A total of eight patients (1 woman and 7 men) with acute bacterial infection were treated with antibiotics therapy, and blood samples were collected during the hospital course. These infections included pneumonia, cellulitis, urinary tract infection, and acute cholecystitis. Different pathogens were growth from their blood culture.

Healthy controls

The control group consisted of 26 healthy donors (18 women and 8 men). These participants were required to complete a brief questionnaire regarding previous or current medical conditions.

Flow cytometric analysis

All SLE patients with or without infection and non-SLE patients with infection were monitored for their plasma cell number in peripheral blood during their hospital stay. In the SLE patient with infection, after the antibiotics or conservative treatment, the percentage of CD27^high plasma cell was checked again 1 to 2 months later. Briefly, 100 µL of EDTA anti-coagulated whole blood was incubated with 20 µL fluorescein isothiocyanate (FITC)-labeled anti-CD19 and 20 µL peridinin chlorophyll protein (Per CP)-labeled anti-CD20, in combination with 20 µL phycoerythrin (PE)-labeled anti-CD27. Tubes were incubated for 20 min at room temperature before lysing solution was added. Erythrocytes were lysed with lysing solution for 10 min in the dark. Blood samples were washed with phosphate-buffered saline and analyzed on a FACS-track cytometer (Becton Dickinson). Lymphocytes were gated on the basis of forward (FSC) and side scatter (SSC). A live gate was set on the CD19+ cell population, and then we evaluated the expression of CD27 and CD20. Frequencies of CD20-CD27+ plasma cells were analyzed and calculated on the same day, using CellQuest software (Becton Dickinson, San Jose, CA, USA).

Statistical analysis

SPSS version 15.0 software (SPSS Inc., Chicago, IL, USA) was used to perform the statistical analyses in our study. Differences between median values of defined patient groups were compared using the nonparametric Mann-Whitney U test. A Spearman's rank correlation was applied, to detect correlations among different study parameters. Any p value less than 0.05 was considered statistically significant.

RESULTS

Characteristics of the four study groups

The study population consisted of 36 SLE patients without infection, 23 SLE patients with infection, eight non-SLE patients with infection, and 26 healthy controls. Among the patients with SLE, 36 patients (33 women and 3 men, mean age 38.3 ± 2.6 years) without infection did not receive any antibiotic therapy, and 23 patients (19 women and 4 men, mean age 39.3 ± 3.4 years) with evidence of viral or bacterial infection received antibiotic therapy (Table 1). In the group of non-SLE patients with infection, the bacterial pathogens are shown in Table 2. General data vis-à-vis the SLE patients of this study are shown in Table 3. SLE patients without infection had significantly higher anti-dsDNA and SLEDAI values and lower CRP and C4 values, compared to SLE patients with viral or bacterial infections. However, the other parameter cannot distinguish lupus flares and infection in SLE patients.

Frequencies of peripheral CD27^high plasma cells in controls, SLE patients, and non-SLE patients with infection

The frequency of peripheral CD27^high plasma cells increased in patients of SLE without infection, SLE with infection, and non-SLE with infection, compared to healthy donors (p < 0.05). There was a significant difference of percentage in circulating CD27^high plasma cells, between the SLE patients with infection and those without infection (p < 0.05). The mean percentage of circulating CD27^high plasma cells was higher in the patients of SLE with infection and in the group of non-SLE with infection, and there was no significant difference between these two groups (Fig. 1). There was no significant difference between SLE patients with bacterial and viral infection (41.3 ± 7.6 vs. 47.4 ± 11.8, p = 0.548). After antibiotic therapy, the percentage of CD27^high plasma cells decreased significantly in the group of SLE patients with infection (43.1 ± 6.3 decreased to 14.1 ± 1.2, p < 0.05). The mean percentage of CD27^high plasma cells and CRP levels of the study group is shown in Table 4. There was a significant correlation between the percentage of peripheral CD27^high plasma cells and disease activity, including SLEDAI (r = 0.866, p < 0.001), anti-dsDNA (r = 0.886, p < 0.001), C3 (r = - 0.392, p = 0.018), and C4 (r = - 0.337, p = 0.044) in the group of SLE without infection; however, there was no statistical correlation with anti-Ro, anti-La, anticardiolipin IgG/IgM, anti-RNP, anti-Sm, or RF. In the SLE patients with infection, there was no correlation between SLE disease activity and circulating CD27^high plasma cells percentage, by SLEDAI or these autoantibodies (Table 5) (Fig. 2).

DISCUSSION

In humans, B cells are important to the immune system, as they produce antibodies, act as antigen-presenting cells, produce cytokines, and interact with T cells to regulate immunity-related activities.2,15 There are many regulatory mechanisms-including apoptosis, receptor editing, anergy, and clonal ignorance-underpinning B cell tolerance in healthy individuals.16-18 The loss of this tolerance may induce autoreactive B cell processes, in turn possibly incurring an autoimmune disorder. In SLE, multiple internal or external factors increase B cell reactivity, so that it can directly result in the production of pathologic autoantibodies and in end-organ damage. There are generally a large number of antibodies from the SLE patient's blood that counter self-antigens. Plasma cells containing surface markers of CD19+CD20- prodigiously secrete antibodies.11 The CD27 molecule belongs to the tumor necrosis factor receptor family and is expressed on T and B cells.11,19 CD27+ B cells are quickly activated to produce a higher level of immunoglobulin in comparison to CD27- B cells.19

Previous studies show that the expansion of CD27^high plasma cells is associated with the serological presence of particular autoantibodies, including anti-dsDNA, anti-Ro, anti-La, anti-Sm, and anti-histone antibodies.10-12 In our data, circulating CD27^high plasma cells seem to be important for the survey of disease activity in SLE patients without infection, because there is a significant correlation between the frequency of CD27^high plasma cells in peripheral blood and disease activity (i.e., SLEDAI, anti-dsDNA, C3 and C4 level, Table 5). However, we cannot use these parameters to evaluate disease activity in SLE patients with infection, due to the fact there is no correlation with disease activity (Table 5).

In SLE patients with or without infection, a significantly increased percentage of CD27^high plasma cells from the peripheral blood can be found from our data (but a significantly higher percentage of CD27^high plasma cells in SLE patients with infection), when compared to healthy controls (Fig. 2). Non-SLE patients with infection also have a higher percentage of CD27^high plasma cells. Infection is an important issue in SLE patients, because it is one of the most frequent causes of death among afflicted individuals.20 CRP is a traditional serological parameter used to distinguish a lupus flare-up from infection.13,14 However, no elevated level of CRP was found in patients with SLE, and the CRP level can be elevated in a lupus flare-up without infection.21,22 For this reason, CRP is not always a good parameter to evaluate SLE patients with or without infection. Novel parameters to differentiate infectious disease from exacerbation in lupus patients include low soluble Fc gamma receptor III; elevated granulocyte colony-stimulating factor; and elevations in sCD14, sICAM-1, or sE-selectin.23,24 In our data, there was no significant difference between SLE patients with infection and non-SLE patients with infection, in terms of the expansion of peripheral CD27^high plasma cells. Infection is associated with immune hyperactivation caused by a triggering of external antigens and immune responses of SLE with flare-up are activated by autoantigens. Under either of these two conditions, hyperactive CD27^high plasma cells increase in number. Infection should be considered another cause of increase in CD27^high plasma cells.12 Plasma cells play a major role to produce antibody by homing to the bone marrow.18 During acute inflammation or infection, transiently elevation of plasma cells can be found. Therefore, the percentage of CD27^high plasma cells can not be used to differentiate SLE patients with infection or non-SLE patients with infection. From our data, there were increasing plasma cell numbers in both infection and SLE flares; however, an increased percentage of CD27^high plasma cells was found in the infection compared with SLE flares. The mechanism was still unclear.

Because of their long-term use of immunosuppressive therapy and persistent hypocomplementaemia, patients with SLE are at higher risk of contracting common and opportunistic infections, compared to the general population.25,26 If SLE patients hold the baseline data of the percentage of circulating CD27^high plasma cells, we can evaluate whether a SLE patient is in an infectious state or not. Higher percentages of CD27^high plasma cells are found among SLE patients with infection, and lower percentages of CD27^high plasma cells (though higher than that in controls) are more prevalent in SLE patients with active disease. In SLE patients with infection, the percentage of CD27^high plasma cells decreased significantly following adequate antibiotics therapy (Fig. 3).

Various microorganisms-including bacterial, viral, fungal, and protozoa infections-are major causes of morbidity and mortality in SLE patients with infection.20,27,28 Timely antibiotic therapy is required for SLE patients with infection, but adequate immunosuppression is required for the control of SLE with flare-up. The percentage of CD27^high plasma cells provides clues for decision-making vis-à-vis appropriate therapy, when active lupus cannot be clearly distinguished from infection.

In summary, the percentage of CD27^high plasma cells from peripheral blood is a good biomarker for evaluating disease activity in SLE patients who do not have infection, and it can also serve as a potential parameter for distinguishing a lupus flare-up from infection.

Figures and Tables

Fig. 1

Comparison of the percentage of overall CD27^high plasma cells from four groups (SLE, SLE with infection, Non-SLE with infection, healthy control). SLE, systemic lupus erythematosus; HC, healthy control.

Fig. 2

Different expressions of peripheral CD27^high plasma cells in the four patient groups. (A) SLE with pulmonary haemorrhage (18%). (B) SLE with Streptococcus pneumonia infection (63%). (C) Non-SLE with Staphylococcus aureus infection (54%). (D) Healthy control (2%). SLE, systemic lupus erythematosus

Fig. 3

In an SLE patient with Streptococcus pneumonia infection, after the antibiotics therapy, there was a significant decline in the percentage of CD27^high plasma cells one month later (33% to 11%). The high level of CRP decreased to normal range (4.2 mg/dL to 0.31 mg/dL). SLE, systemic lupus erythematosus; CRP, C-reactive protein.

Table 1

Clinical Pathogens and Characteristics of Systemic Lupus Erythematosus Patients with Bacterial and Viral Infection

Table 2

Clinical Characteristics of Non-Systemic Lupus Erythematosus Patients with Bacterial Infection

Table 3

Clinical Characteristics and Serological Parameters of 36 Systemic Lupus Erythematosus (SLE) Patients and 23 SLE Patients with Infection

SLEDAI, Systemic Lupus Erythematosus Disease Activity Index; CRP, C-reactive protein; BUN, Blood urea nitrogen; Cr, Creatinine; NS, not stignificant.

Normal rage of anti-dsDNA (0 - 10), C3 (90 - 180), C4 (10 - 40), CRP (0 - 0.5), White blood cell (4,500 - 11,000), Hemoglobin (12 - 16), Platelet (150,000 - 400,000), BUN (7 - 20), Cr (0.5 - 1).

The values are presented as mean ± SEM.

Table 4

Mean Percentage of CD27^high Plasma Cells in Peripheral Blood and Serum C-reactive Protein Levels, of All Studied Patients

HC, healthy control; CRP, C-reactive protein; SLE, systemic lupus erythematosus.

^*Percentages of CD27^high plasma cells are significantly higher in three groups (SLE, SLE with infection, non-SLE with infection), when compared to healthy controls (p < 0.05).

^†Mean CRP levels are higher in SLE with infection and non-SLE with infection, when compared to SLE (p < 0.05).

^‡Mean percentage of CD27^high plasma cells is significantly lower in SLE, when compared to SLE with infection (p < 0.05).

Table 5

The Correlation between the Percentage of CD27^high Plasma Cells and the Marker of Systemic Lupus Erythematosus Disease Activity

SLEDAI, Systemic Lupus Erythematosus Disease Activity Index.

ACKNOWLEDGEMENTS

This study was supported by Tri-Service General Hospital and National Defense Medical Center (Grant no: TSGHC99-141).

Notes

The authors have no financial conflicts of interest.

References

1. Chan OT, Madaio MP, Shlomchik MJ. The central and multiple roles of B cells in lupus pathogenesis. Immunol Rev. 1999. 169:107–121.

2. Lipsky PE. Systemic lupus erythematosus: an autoimmune disease of B cell hyperactivity. Nat Immunol. 2001. 2:764–766.

3. Odendahl M, Jacobi A, Hansen A, Feist E, Hiepe F, Burmester GR, et al. Disturbed peripheral B lymphocyte homeostasis in systemic lupus erythematosus. J Immunol. 2000. 165:5970–5979.

4. Wei C, Anolik J, Cappione A, Zheng B, Pugh-Bernard A, Brooks J, et al. A new population of cells lacking expression of CD27 represents a notable component of the B cell memory compartment in systemic lupus erythematosus. J Immunol. 2007. 178:6624–6633.

5. Jacobi AM, Reiter K, Mackay M, Aranow C, Hiepe F, Radbruch A, et al. Activated memory B cell subsets correlate with disease activity in systemic lupus erythematosus: delineation by expression of CD27, IgD, and CD95. Arthritis Rheum. 2008. 58:1762–1773.

6. Han BK, White AM, Dao KH, Karp DR, Wakeland EK, Davis LS. Increased prevalence of activated CD70+ CD4+ T cells in the periphery of patients with systemic lupus erythematosus. Lupus. 2005. 14:598–606.

7. Dolff S, Wilde B, Patschan S, Dürig J, Specker C, Philipp T, et al. Peripheral circulating activated b-cell populations are associated with nephritis and disease activity in patients with systemic lupus erythematosus. Scand J Immunol. 2007. 66:584–590.

8. Liu CC, Ahearn JM. The search for lupus biomarkers. Best Pract Res Clin Rheumatol. 2009. 23:507–523.

9. Yang DH, Chang DM, Lai JH, Lin FH, Chen CH. Usefulness of erythrocyte-bound C4d as a biomarker to predict disease activity in patients with systemic lupus erythematosus. Rheumatology (Oxford). 2009. 48:1083–1087.

10. Jacobi AM, Odendahl M, Reiter K, Bruns A, Burmester GR, Radbruch A, et al. Correlation between circulating CD27 high plasma cells and disease activity in patients with systemic lupus erythematosus. Arthritis Rheum. 2003. 48:1332–1342.

11. Dörner T, Lipsky PE. Correlation of circulating CD27 high plasma cells and disease activity in systemic lupus erythematosus. Lupus. 2004. 13:283–289.

12. Ten Boekel E, Siegert CE, Vrielink GJ, Van Dam VC, Ceelen A, De Kieviet W. Analyses of CD27++ plasma cells in peripheral blood from patients with bacterial infections and patients with serum antinuclear antibodies. J Clin Immunol. 2007. 27:467–476.

13. Suh CH, Jeong YS, Park HC, Lee CH, Lee J, Song CH, et al. Risk factors for infection and role of C-reactive protein in Korean patients with systemic lupus erythematosus. Clin Exp Rheumatol. 2001. 19:191–194.

14. ter Borg EJ, Horst G, Limburg PC, van Rijswijk MH, Kallenberg CG. C-reactive protein levels during disease exacerbations and infections in systemic lupus erythematosus: a prospective longitudinal study. J Rheumatol. 1990. 17:1642–1648.

15. Fritsch RD, Shen X, Illei GG, Yarboro CH, Prussin C, Hathcock KS, et al. Abnormal differentiation of memory T cells in systemic lupus erythematosus. Arthritis Rheum. 2006. 54:2184–2197.

16. Swaak AJ, Hintzen RQ, Huysen V, van den Brink HG, Smeenk JT. Serum levels of soluble forms of T cell activation antigens CD27 and CD25 in systemic lupus erythematosus in relation with lymphocytes count and disease course. Clin Rheumatol. 1995. 14:293–300.

17. Font J, Pallares L, Martorell J, Martinez E, Gaya A, Vives J, et al. Elevated soluble CD27 levels in serum of patients with systemic lupus erythematosus. Clin Immunol Immunopathol. 1996. 81:239–243.

18. Carter RH. B cells in health and disease. Mayo Clin Proc. 2006. 81:377–384.

19. Agematsu K, Hokibara S, Nagumo H, Komiyama A. CD27: a memory B-cell marker. Immunol Today. 2000. 21:204–206.

20. Cervera R, Khamashta MA, Font J, Sebastiani GD, Gil A, Lavilla P, et al. Morbidity and mortality in systemic lupus erythematosus during a 10-year period: a comparison of early and late manifestations in a cohort of 1,000 patients. Medicine (Baltimore). 2003. 82:299–308.

21. Roy S, Tan KT. Pyrexia and normal C-reactive protein (CRP) in patients with systemic lupus erythematosus: always consider the possibility of infection in febrile patients with systemic lupus erythematosus regardless of CRP levels. Rheumatology (Oxford). 2001. 40:349–350.

22. Williams RC Jr, Harmon ME, Burlingame R, Du Clos TW. Studies of serum C-reactive protein in systemic lupus erythematosus. J Rheumatol. 2005. 32:454–461.

23. Hellmich B, Csernok E, de Haas M, von dem Borne AE, Schatz H, Gross WL, et al. Low Fcgamma receptor III and high granulocyte colony-stimulating factor serum levels correlate with the risk of infection in neutropenia due to Felty's syndrome or systemic lupus erythematosus. Am J Med. 2002. 113:134–139.

24. Egerer K, Feist E, Rohr U, Pruss A, Burmester GR, Dörner T. Increased serum soluble CD14, ICAM-1 and E-selectin correlate with disease activity and prognosis in systemic lupus erythematosus. Lupus. 2000. 9:614–621.

25. Zandman-Goddard G, Shoenfeld Y. Infections and SLE. Autoimmunity. 2005. 38:473–485.

26. Bosch X, Guilabert A, Pallarés L, Cerveral R, Ramos-Casals M, Bové A, et al. Infections in systemic lupus erythematosus: a prospective and controlled study of 110 patients. Lupus. 2006. 15:584–589.

27. Zandman-Goddard G, Shoenfeld Y. SLE and infections. Clin Rev Allergy Immunol. 2003. 25:29–40.

28. Gladman DD, Hussain F, Ibañez D, Urowitz MB. The nature and outcome of infection in systemic lupus erythematosus. Lupus. 2002. 11:234–239.