Elevation of inflammatory markers in patients with systemic lupus erythematosus is associated with poorer outcome

Biomedicine & Pharmacotherapy 67 (2013) 48–52

Available online at

www.sciencedirect.com

Original article

Elevation of inflammatory markers in patients with systemic lupus erythematosus is associated with poorer outcome Anat Gafter-Gvili a,c,*, Leonard Leibovici a,c, Yair Molad b,c a

Department of Medicine E, Rabin Medical Center, Petah Tikva, Israel Rheumatology Unit, Rabin Medical Center, Petah Tikva, Israel c Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 24 August 2012 Accepted 22 October 2012

Objective: To determine the association between inflammatory markers and mortality in patients with systemic lupus erythematosus (SLE). Methods: A retrospective cohort analysis of 143 patients with SLE followed between 1991 and 2010 in a Lupus Clinic in Israel. High sensitivity CRP (hsCRP) level and ESR were recorded at baseline. We compared outcomes of patients with elevation of either ESR or hsCRP to outcomes of patients without any elevation of inflammatory markers at start of follow-up. Risk factors for mortality were identified using univariate and multivariate analyses (Cox regression analysis). Results: Of the 143 patients, 93 patients had an inflammatory marker and 50 had no inflammatory marker. There were no differences between the two groups in terms of lupus criteria, disease activity parameters or damage index. Survival was lower for patients in the inflammatory group, 24 deaths of 93 patients (25.8%), 1393 patients-years, versus five deaths of 50 patients (10%), 692 patient-years, in the non-inflammatory group, log rank P = 0.031. On multivariate analysis, inflammatory markers at baseline remained an independent risk factor for death, hazard ratio 2.72 (95% CI, 1.3–7.2). Conclusions: SLE patients with elevation of inflammatory markers at baseline are at higher risk of death. ß 2012 Elsevier Masson SAS. All rights reserved.

Keywords: C-reactive protein Erythrocyte sedimentation rate Systemic lupus erythematosus Mortality SLICC/ACR damage index

1. Introduction Systemic lupus erythematosus (SLE) is an autoimmune disease in which inflammation and tissue damage are predominant features [1], thus markers of acute-phase reaction may be elevated throughout its course. Two common methods for detecting the acute-phase response of an inflammatory reaction are the erythrocyte sedimentation rate (ESR) and the more specific and sensitive C-reactive protein (CRP) [2]. In patients with SLE, elevated ESR has been found to be associated with both disease activity and damage accrual [3], whereas serum CRP is only modestly elevated and is not correlated with disease activity [4,5]. Although serum level of CRP may remain within normal range during lupus flare-up, it is still higher than in healthy control subjects [6]. It has been consistently shown that disease activity and damage scored by the Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology Damage (ACR) damage index (DI) are powerful predictors of poor outcome in patients with SLE [7]. In recent years, it has been hypothesized that inflammation * Corresponding author. Department of Medicine E, Rabin Medical Center, Petah Tikva 49100, Israel. Tel.: +972 3 9376500; fax: +972 3 9376512. E-mail addresses: [email protected], [email protected] (A. Gafter-Gvili). 0753-3322/$ – see front matter ß 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.biopha.2012.10.009

plays a pivotal role in the initiation and progression of vascular atherosclerosis [8]. Moreover, CRP has also been linked to mortality in apparently healthy individuals [9]. Therefore, we sought to assess the association of elevated inflammatory markers (ESR and CRP) with mortality in a singlecenter cohort of patients with SLE. 2. Methods 2.1. Setting and patients This retrospective cohort study includes SLE patients under routine follow-up (every 3 to 4 months) during a 19-year period (1991 to 2010) at the Lupus Clinic of Rabin Medical Center, Petah Tikva, Israel, which serves as a university-affiliated primary community hospital as well as a referral tertiary medical center. All patients included fulfilled at least four of the American College of Rheumatology 1982 revised criteria for SLE [10]. 2.2. Inflammatory markers and definitions High sensitivity CRP (hsCRP) and ESR were measured at start of follow-up. hsCRP is assayed by our hospital’s biochemistry laboratory using the immunoturbidimetric Tina-quant CRP

A. Gafter-Gvili et al. / Biomedicine & Pharmacotherapy 67 (2013) 48–52

method (Roche Hitachi 911) (normal range 0.00–0.50 mg/dl). ESR is measured in our hospital’s hematology laboratory by the Sedimatic 100 analyzer (Pharmatop). We divided our cohort into two groups: the first including patients with no elevation of an inflammatory marker and the second including patients with either elevation of hsCRP, ESR or both. Elevation of CRP was considered if the value was above the upper normal value (> 0.5 mg/dl) and elevation of ESR was considered if greater than 40 mm/h. 2.3. Data collection Data from medical charts were recorded and analyzed. Additional information was obtained from the hematological, biochemical and immunological laboratories of the hospital. For purposes of the study, the visit at which hsCRP level or ESR level was first recorded was considered the initial visit. If at the first encounter one of them was missing, we obtained hsCRP level or

49

ESR within 4 months. For each patient we recorded outcome data available at the latest follow-up. At each encounter, the patient’s medical history was recorded and a physical examination was performed. All data were immediately converted by the examining physician into 25 medical history and 29 physical examination variables [11]. Additional blood tests at baseline were performed and recorded: complete blood count, serum double-stranded (ds) DNA (by Farr assay) and complement (C3, C4, CH100) testing. We recorded data regarding atherosclerotic risk factors and comorbidities: smoking, hypertension, diabetes mellitus, hypercholesterolemia, coronary artery disease (CAD), stroke, congestive heart failure (CHF), and malignancy. Disease activity was retrospectively scored using the SLE Disease Activity Index (SLEDAI) [12] and disease-related end-organ damage was retrospectively scored with the SLICC/ACR DI [13], according to the data from the initial visit. Number of deaths and the causes of death were ascertained by a review of patient files, and by the hospital computerized data

Table 1 Disease characteristics, comorbidities and therapy at initial visit in 143 patients with SLEa,b. All patients (n = 143)

Patients with inflammatory markers (n = 93)

Patients with no inflammatory markers (n = 50)

Demography, laboratory Age (years), median (range) Female n/n (%) Number of SLE criteria, median (range) CRP mg/dl, median (range) ESR mm/h, median (range) Hbg/dl median (range) WBC (mm3), median (range) Platelets (mm3), median (range) Anti-ds DNA antibody (%), rangec C3 (mg/dl), rangec C4 (mg/dl), rangec CH100 (%), rangec

41 (15–79) 132 (92.3) 4.95 (1.165) 0.5 (0–31.8) 40 (0–134) 12.4 (8.5–13.42) 6500 (100–26,800) 230,000 (28,000–638,000) 22.5 (0–120) 93 (25–186) 20 (5–107) 61 (0–180)

32 (4–70) 87/93 (93.5) 5 (0–9) 0.9 (0–31.8) 60 (1–134) 12.2 (8.5–13.05) 6580 (100–19,800) 236,000 (41,000–638,000) 26.5 (0–120) 92 (25–186) 20 (6–107) 65 (0–180)

38 (18–79) 45/50 (90) 5 (4–7) 0.14 (0–0.5) 19 (0–36) 12.7 (9–13.42) 6400 (2100–26,800) 216,000 (28,000–510,000) 19.5 (0–93) 95 (26–163) 20 (5–44) 55 (16–100)

Disease score (mean, range) SLEDAI SLICC/ACR DI

3 (0–25) 0.92 (0–9)

4 (0–25) 0 (0–9)

2 (0–12) 0 (0–4)

Disease manifestations n/n (%) Malar rash Discoid rash Photosensitivity Oral ulcers Arthritis Serositis Renal disease Neuropsychiatric disease Hematologicgic disease Immunologic criteria ANA

64/143 (44.8) 23/143 (16.1) 39/143 (27.3) 32/143 (22.4) 91/143 (63.6) 45/143 (31.5) 51/143 (35.7) 21/143 (14.7) 67/143 (46.9) 120/143 (83.9) 134/143 (93.7)

43/93 13/93 24/93 20/93 62/93 34/93 32/93 15/93 38/93 78/93 88/93

(46.2) (14) (25.8) (21.5) (66.7) (26.6) (34.4) (16.1) (40.9) (83.9) (94.6)

21/50 (42) 10/50 (20) 15/50 (30) 12/50 (24) 29/50 (58) 11/50 (22) 19/50 (38) 6/50 (12) 29/50 (58) 42/50 (84) 46/50 (92)

Atherosclerotic risk factors/manifestations Diabetes mellitus Hypertension Smoking Hypercholesterolemia Coronary artery disease Congestive heart failure CVA TIA Venothromboembolic disease Malignancy

12 (8.4) 49 (34.3) 18 (12.6) 25 (17.5) 7 (4.9) 3 (2.1) 8 (5.6) 4 (2.8) 21 (14.7) 2 (1.4)

8/93 (8.6) 30/93 (32.3) 15/93 (16.1)

4/50 (8) 19/50 (38) 3/50 (6)

4/93 (4.3) 2/93 (2.2) 4/93 (4.3) 3/93 (3.2) 12/93 (12.9) 1/93 (1.1)

3/50 1/50 4/50 1/50 9/50 1/50

Medications Steroid therapy Cytotoxic drugs

72 (50.3) 47 (32.9)

46/93 (49.5) 33/93 (38.7)

26/50 (52) 11/50 (22)

(6) (2) (8) (2) (18) (2)

a ANA: antinuclear antibody; CVA: cerebrovascular accident; SD: standard deviation; SLEDAI: SLE disease activity index; SLICC/ACR: Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage damage index; TIA: transient ischemic attack. b Continuous variables described as medians with range, compared using the Mann-Whitney U test. All of the comparisons were not statistically significant, P > 0.05, except cytotoxic drugs, P = 0.04. c Normal range: anti-ds DNA: 0–20%; C3: 88–201 mg/dl; C4: 16–47 mg/dl; CH100: 40–100%.

50

A. Gafter-Gvili et al. / Biomedicine & Pharmacotherapy 67 (2013) 48–52

system and death certificates. In addition, survival status was ascertained using the central registry of the Israeli Ministry of Internal Affairs. 2.4. Statistical analysis We primarily assessed the association between inflammatory markers and survival since start of follow-up. Specifically, we compared outcomes of patients with elevation of either ESR or hsCRP to outcomes of patients without any inflammatory markers at baseline. Dichotomous variables were examined by Chi-square test or Fisher’s exact test. Continuous variables were described as medians and range, and compared with the Mann-Whitney U test. P values less than or equal to 0.05 were considered statistically significant. Time to death was compared using Kaplan-Meier survival analysis. Independent risk factors for early death were identified using Cox regression analysis, including variables associated with mortality on univariate analysis (P < 0.05). Hazard ratios (HRs) for death with 95% confidence intervals (CIs) were calculated and reported. Statistical analysis was performed using IBM1 SPSS1 Statistic 19 (IBM Corporation, NY).

3. Results 3.1. Characteristics of patients One hundred and forty-three patients with SLE were recruited during the study period, from 1991 until 2010. Mean follow-up period was 9.4 years (SD 3.3) and median was 9.0 (range 1–19), translating into 1344 patient-years of follow-up. The cohort included 132 (92.3%) women and 11 men (7.7%) who fulfilled the American College of Rheumatology 1982 revised criteria for SLE, with a mean criteria number of 5.0 (SD 1.2). Median age at study entry was 41 (range 15– 79) years. Of the 143 patients, 93 patients had an elevated inflammatory marker and 50 had an inflammatory marker within the normal limit. Table 1 depicts the disease characteristics, comorbidities, and therapies used, as recorded at the initial encounter. There were no statistically significant differences between the two groups in terms of lupus criteria, except a trend for more serositis in the inflammatory group (34/93 patients in the inflammatory group, 36.6% vs. 11/50 patients in the noninflammatory group, 22%, P = 0.07). There was also no difference in disease activity parameters including SLEDAI score, serum antids DNA antibody and complement level, as well as the SLICC/ACR DI, clinical manifestations of SLE and comorbidities (Table 1). The only difference was higher rate of use of cytotoxic drugs in the inflammatory group 33/93 (38.7%) vs. 11/50 (22%) in the noninflammatory group, P = 0.04.

Fig. 1. Kaplan-Meier analysis: time to death for patients with SLE, with and without inflammatory markers at baseline.

10%), 692 patient-years, in the non-inflammatory group, Log rank P = 0.031 (Fig. 1). Other variables significantly associated with death on univariate analysis were older age at study entry, SLICC/ACR DI and male sex. Data for older age at study entry: survival for 1st quartile, four deaths of 34 patients (11.8%), for 2nd quartile, five deaths of 39 (12.8%), for 3rd quartile, six deaths of 35 patients (17.1%) and for 4th quartile, 14 deaths of 35 patients (40%), P = 0.003. Death was also significantly associated with higher SLICC/ACR DI at baseline: for 1st percentile, seven deaths of 79 patients (9.9%), for 2nd percentile, six deaths of 28 (21.4%), and for 3rd percentile, 16 deaths of 36 patients (54.4%), P < 0.0001. Male sex was positively correlated with higher risk of death: five deaths of 11 male patients (55.5%) versus 24 deaths of 132 female patients (18.2%), P = 0.014. However, baseline SLEDAI score was not found to be significantly associated with death (P = 0.517). On Cox regression multivariable analysis, elevated baseline inflammatory markers remained an independent risk factor for death, HR 2.72 (95% CI, 1.3–7.2). SLICC/ACR DI was also independently associated with death, HR 1.28 (95% CI, 1.08– 1.50) for 1 index point. Female sex was independently associated with improved survival, HR 0.37 (95% CI, 0.14–1.00).

3.2. Mortality Twenty-nine (20.3%) patients died during the study follow-up period. Causes of death were as follows: sepsis (8 patients, 5.6% of all patients, 27.6% of all mortality cases), malignancy (4 patients, 2.3% of all patients, 13.8% of all mortality cases), stroke (3 patients, 2.1% of all patients, 10.34% of all mortality cases), cardiac causes (2 patients, 1.55% of all patients, 6.9% of all mortality cases), death from unknown causes (12 patients, 8.4% of all patients, 27.6% of all mortality cases). Survival at the end of follow-up was lower for patients in the inflammatory group, 24 deaths of 93 patients (mortality of 25.8%), 1393 patients-years, versus five deaths of 50 patients (mortality of

4. Discussion We present data from a cohort of 143 SLE patients in a single center in Israel, with a long follow-up. In this retrospective analysis, we show that baseline inflammatory markers within the normal range are predictive for improved survival and lower risk for death, while elevated inflammatory markers (either hsCRP or ESR) independently predict decreased survival. Lack of elevated inflammatory markers remained an independent predictor for better survival in a multivariable analysis, as did SLICC/ACR DI and sex.

A. Gafter-Gvili et al. / Biomedicine & Pharmacotherapy 67 (2013) 48–52

Moreover, the groups of patients with or without elevated inflammatory markers did not differ in other characteristics of SLE or comorbidities. These findings suggest that elevation of inflammatory markers may have a role in the long-term outcome of SLE. Our results are in concert with former reports. Disease activity and overall damage as measured by SLICC/ACR DI have been reported as powerful predictors of poor survival in patients with SLE [7]. Other studies have also shown that the accrual of damage in SLE is predicted by disease activity [14]. We showed that both SLICC/ACR DI and an elevated marker of inflammation correlate with mortality. Elevated ESR alone has previously been linked to disease activity and damage accrual in SLE in a large multiethnic cohort of 513 SLE patients participating in the LUMINA study (Lupus in minority Populations: Nature versus Nurture). Moderate and mild elevations of ESR were independently associated with SLICC/ACR DI [3]. CRP in SLE population was also found to be associated with damage. In 610 patients participating in the Hopkins Lupus Cohort prospective study, hsCRP was found to be strongly associated with total SLICC/ACR DI as well as with disease activity [15]. Moreover, elevated CRP was found to be associated with death in the general population. In a meta-analysis of 160,309 people with no history of cardiovascular disease from 54 long-term prospective studies, CRP concentration was associated with the risk of CAD and non-vascular mortality [9]. The same relationship in lupus can probably be explained by a common acute-phase response to the inflammatory, infectious, atherosclerotic and malignant diseases which precede death in both lupus and nonlupus patients. Inflammation appears to be a direct risk factor for atherosclerosis [3]. This assumption is based on studies showing that CRP is a powerful, independent predictor of CAD, stroke, and vascular death in the general population, regardless of ethnicity, sex, or age [9]. Evidence suggests that enhanced atherosclerosis may cause premature cardiovascular events in SLE [16]. Moreover, there is evidence to support hsCRP as an independent predictor of vascular events in patients with SLE [5]. Several studies found an association between CRP and intima-media wall thickness, which are predictors of future cardiovascular events and mortality, in SLE patients [17,18]. Another prospective multiethnic SLE cohort study supporting the use of hsCRP in lupus patients showed that, in Caucasians, the proportion of patients with hsCRP values in the highest quintile was significantly higher among those who developed vascular events than among those who did not [19]. We hypothesize that patients who have increased inflammation as reflected by elevated inflammatory markers, are possibly at an increased risk for accelerated atherosclerosis, although this was not evaluated in our study. The strength of our results is that they may be applicable to all SLE patients, since our cohort represents a typical SLE cohort. The most common SLE manifestations were arthritis, rash and hematological abnormalities, as previously described [1]. We demonstrated a mortality rate of 20.3%, during a 19-year follow-up period, which is consistent with other reports [20,21]. The distribution of causes of death is also consistent with that reported in the literature, with infections, vascular causes (both cardiovascular and cerebrovascular) and malignancy being the most common causes for death [20–23]. Another advantage is the long-term follow-up of almost 20 years. 4.1. Limitations Our results are limited by the retrospective design. In addition, it is a small cohort with a relatively small number of events. We only have data regarding baseline ESR and CRP levels, and these might change over time. Moreover, ESR and CRP are not specific to

51

SLE, and we did not ascertain other coexistent conditions at baseline that may be associated with elevated inflammatory markers. In addition, our study was not powered enough to examine the interrelations between inflammation and cardiovascular death. 4.2. Implication for practice Despite the limitations our results have clinical implications. Our study suggests that elevation of inflammatory markers at baseline has prognostic significance. Our results emphasize the prognostic value of such cheap and easy to perform laboratory parameters such as ESR and hsCRP for the assessment of disease outcome, albeit being non-specific in SLE. 4.3. Implication for research Our findings should be validated in long-term prospective studies and we suggest that baseline ESR and hsCRP values should be incorporated in future studies in SLE. Future studies should assess inflammatory markers both at baseline and also during follow-up. Elevation of ESR and CRP may serve as a platform for designing trials of interventions in SLE. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Funding statement: this research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. References [1] Tsokos GC. Systemic lupus erythematosus. N Engl J Med 2011;365(22): 2210–21. [2] Osei-Bimpong A, Meek JH, Lewis SM. ESR or CRP? A comparison of their clinical utility. Hematology 2007;12(4):353–7. [3] Vila´ LM, Alarco´n GS, McGwin Jr G, Bastian HM, Fessler BJ, Reveille JD. Systemic lupus erythematosus in a multiethnic cohort (LUMINA): XXIX. Elevation of erythrocyte sedimentation rate is associated with disease activity and damage accrual. J Rheumatol 2005;32(11):2150–5. [4] Gaitonde S, Samols D, Kushner I. C-reactive protein and systemic lupus erythematosus. Arthritis Rheum 2008;59(12):1814–20. [5] de Carvalho JF, Hanaoka B, Szyper-Kravitz M, Shoenfeld Y. C-reactive protein and its implications in systemic lupus erythematosus. Acta Reumatol Port 2007;32(4):317–22. [6] Hasselink DA, Aardein LA, Swaak AJ. Profiles of the acute-phase reactants Creactive protein and ferritin related to the disease course of patients with systemic lupus erythematosus. Scand J Rheumatol 2003;32:151–5. [7] Kasitanon N, Magder LS, Petri M. Predictors of survival in systemic lupus erythematosus. Medicine (Baltimore) 2006;85(3):147–56. [8] Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med 1999;340: 115–26. [9] Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, Collins R, et al. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: an individual participant meta-analysis. Lancet 2010;375:132–40. [10] Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7. [11] Molad J, Gorshtein A, Wysenbeek AJ, Guedj D, Majadla R, Weinberger A, et al. Protective effect of hydroxychloroquine in systemic lupus erythematosus. Prospective long-term study of an Israel cohort. Lupus 2002;11:356–61. [12] Gladman DD, Ibanez D, Urowitz MB. Systemic Lupus Erythematosus Disease Activity Index 2000. J Rheumatol 2002;29:288–91. [13] Gladman DD, Urowitz MB, Goldsmith CH, Fortin P, Ginzler E, Gordon C, et al. The reliability of the Systemic Lupus Collaborating Clinics/American College of Rheumatology Damage Index in patients with systemic lupus erythematosus. Arthritis Rheum 1997;40:809–13. [14] Becker-Merok A, Nossent HC. Damage accumulation in systemic lupus erythematosus and its relation to disease activity and mortality. J Rheumatol 2006;33(8):1570–7. [15] Lee SS, Singh S, Link K, Petri M. High-sensitivity C-reactive protein as an associate of clinical subsets and organ damage in systemic lupus erythematosus. Semin Arthritis Rheum 2008;38(1):41–54.

52

A. Gafter-Gvili et al. / Biomedicine & Pharmacotherapy 67 (2013) 48–52

[16] Gasparyan AY, Stavropoulos-Kalinoglou A, Mikhailidis DP, Toms TE, Douglas KM, Kitas GD. The rationale for comparative studies of accelerated atherosclerosis in rheumatic diseases. Curr Vasc Pharmacol 2010;8(4):437–49. [17] Doria A, Shoenfeld Y, Wu R, Gambari PF, Puato M, Ghirardello A, et al. Risk factors for subclinical atherosclerosis in a prospective cohort of patients with systemic lupus erythematosus. Ann Rheum Dis 2003;62:1071–7. [18] Selzer F, Sutton-Tyrrell K, Fitzgerald SG, Pratt JE, Tracy RP, et al. Comparison of risk factors for vascular disease in the carotid artery and aorta in women with systemic lupus erythematosus. Arthritis Rheum 2004;50:151–9. [19] Toloza SM, Uribe AG, McGwin Jr G, Alarco´n GS, Fessler BJ, Bastian HM, et al. Systemic lupus erythematosus in a multiethnic US cohort (LUMINA) XXIII. Baseline predictors of vascular events. Arthritis Rheum 2004;50(12):3947–57.

[20] Moss KE, Ioannou Y, Sultan SM, Haq I, Isenberg DA. Outcome of a cohort of 300 patients with systemic lupus erythematosus attending a dedicated clinic for over two decades. Ann Rheum Dis 2002;61(5):409–13. [21] Funauchi M, Shimadzu H, Tamaki C, Yamagata T, Nozaki Y, Sugiyama M, et al. Survival study by organ disorders in 306 Japanese patients with systemic lupus erythematosus: results from a single center. Rheumatol Int 2007;27(3): 243–9. [22] Bernatsky S, Boivin JF, Joseph L, Manzi S, Ginzler E, Gladman DD, et al. Mortality in systemic lupus erythematosus. Arthritis Rheum 2006;54(8):2550–7. [23] Nossent J, Cikes N, Kiss E, Marchesoni A, Nassonova V, Mosca M, et al. Current causes of death in systemic lupus erythematosus in Europe, 2000–2004: relation to disease activity and damage accrual. Lupus 2007;16(5):309–17.