High-sensitivity C-reactive protein and erythrocyte sedimentation rate in systemic lupus erythematosus

Abstract

Levels of C-reactive protein (CRP) have been shown to rise in acute illnesses such as infections and some autoimmune diseases, but not in flares of systemic lupus erythematosus (SLE). Our goal was to investigate the high-sensitivity CRP (hsCRP) response to infection versus disease flare in patients with SLE, and to compare this with the erythrocyte sedimentation rate (ESR) response in these patients. We aimed to determine the hsCRP level that distinguishes between infection and flare in SLE, and investigated the correlation between hsCRP and organ involvement in SLE. We reviewed electronic medical records of all patients with SLE admitted to Cedars Sinai Medical Center between 28 August 2001 and 27 April 2008. Patients were divided into three groups based on the reason for hospitalization: 1) lupus flare; 2) active infection; and 3) both lupus flare and active infection. Data were collected on patient demographics, medication use, microbial culture results, organ involvement in lupus flare, ESR and CRP levels. Data were collected on 85 eligible patients, of whom 54 had a lupus flare, 22 had active infection and eight had both. While the ESR levels did not differ significantly between patients with disease flare and active infection, the hsCRP level was significantly lower in the lupus flare group than in the infection group. Most patients in the lupus flare group who had a significantly high hsCRP level had serositis. We found that at a cut-off of above 5 mg/dl, hsCRP level was correlated with infection with a specificity of 80%. At a cut-off of above 6 mg/dl, hsCRP correlated with infection with a specificity of 84%. hsCRP level was found to be significantly higher in patients with pulmonary involvement than without. hsCRP levels are significantly lower in SLE patients with disease flare than in those with active infection. Elevated hsCRP levels can be used as a predictor of active infection in SLE patients with a high specificity. We review the relationship between IL-6 and hsCRP production in lupus patients.

Keywords

erythrocyte sedimentation rate high-sensitivity C-reactive protein systemic lupus erythematosus

Introduction

C-reactive protein (CRP) levels rise significantly in infection as well as in many rheumatologic diseases, including rheumatoid arthritis (RA)¹ ^– ⁴ and vasculitis.⁵ In patients with systemic lupus erythematosus (SLE) however, its behavior has been perplexing. Several studies investigating the role of CRP in patients with SLE have concluded that CRP levels rise significantly in SLE patients with active infection; however, CRP elevation is only modest at best in SLE patients with active disease without evidence of infection.^2,6 ^– ¹³

Over the past few years, conventional methods of CRP measurement, typically detecting levels above 3–10 mg/l, have been replaced by newer measurement methods which detect levels as low as 0.2 mg/l. These novel approaches to detecting CRP, or high-sensitivity CRP (hsCRP) assays, have expanded the indications for its use to include the evaluation of conditions thought to be associated with inflammation in otherwise healthy individuals and persons with diseases not felt to be traditionally associated with inflammation. For example, recent studies have investigated the role of elevated levels of hsCRP as a risk factor for cardiovascular disease.^9,14 ^– ¹⁶

The distinction between infection and disease activity in patients with SLE often presents a great challenge for clinicians. Conditions such as pneumonitis, arthritis, and meningitis can be produced by both active SLE and infection, and the distinction between the two is not always clear.⁹ In addition, SLE patients may present with both active disease and infection. Differentiating between the two in lupus patients will help guide the treatment approaches in these patients, yet there has been some controversy surrounding the question whether CRP can be useful in making this distinction. With the institution of more sensitive ways of measuring CRP, it is useful to explore whether hsCRP will be helpful in distinguishing infection versus active disease in patients with SLE.

In the present study we investigated the relationship between levels of hsCRP and erythrocyte sedimentation rate (ESR) with either disease activity or infection in SLE in a large cohort of SLE patients hospitalized at our institution. We also attempted to determine if levels of hsCRP can be used to distinguish between disease exacerbations and infections in lupus when compared with the utility of ESR measurements. Finally, we examined any correlation between the hsCRP levels and organ involvement in SLE flare.

Methods

Patients

We identified all patients with a lupus diagnosis admitted to Cedars Sinai Medical Center between 28 August 2001 and 27 April 2008. The initial date was chosen as the earliest time that rate turbidimetry was used to measure hsCRP in our institution. Patients with a lupus diagnosis were identified using lupus procedure and diagnosis codes from the medical records database. Those patients who had both ESR and hsCRPs levels recorded within the same week were selected. Diagnosis of SLE was confirmed if at least four of the 11 ACR criteria for the classification of SLE were documented in the medical records, or if the patient had a kidney biopsy consistent with lupus nephritis. Patients whose lupus diagnosis could not be confirmed and patients admitted to the hospital for conditions unrelated to lupus, such as elective surgical procedures, were excluded from the study.

The electronic medical records of the eligible patients were reviewed, and data were gathered from admitting history and physical, discharge summaries, and laboratory studies. Data were collected on each patient’s age, sex, ethnicity, medication use, specifically prednisone, plaquenil, methotrexate, mycophenolate mofetil, azathioprine, biological agents, other immunosuppressive agents, statins, and hormone replacement therapy. We recorded the reason for admission, culture results, ESR and hsCRP levels, and levels for hemoglobin and creatinine on the date ESR and hsCRP were collected. For patients who had multiple ESR and hsCRP levels documented, only the first values obtained after the patient’s admission were used. The patients were divided into three groups based on the reason for hospital admission: (1) patients with lupus flare; (2) patients with active infection; and (3) patients with both infection and active lupus flare. The second group was further divided into those where infection was proven by culture and those where active infection was strongly suspected, but cultures were either negative or not available (Figure 1).

Laboratory measurement

ESR was measured using modified Westergren method (Excyte ESR Non-Vacuum Tubes kit, EX-10100, Vital Diagnostics Inc.). hsCRP was measured using rate turbidimetry (IMMAGE CRPH reagent Immunochemistry Systems and Calibrator 5 Plus) at Cedars Sinai Medical Center laboratory.

Statistical analysis

Numerical variables were summarized by means and standard deviations (if normally distributed) or by medians and interquartile ranges (if not normally distributed). Categorical variables were summarized by frequencies and percents. Group differences on normally distributed numerical variables were assessed by the independent samples t-test (Groups 1 and 2) and ANOVA (Groups 1, 2, and 3). Group differences on non-normally distributed numerical variables were assessed by the Wilcoxon rank sum test (Groups 1 and 2) or the Kruskal–Wallis test (Groups 1, 2, and 3). Group differences on categorical variables were assessed by the Chi-Square or Fisher exact test, as appropriate. Logistic regression models were used to assess the predictive value of ESR and hsCRP for the probability of active infection. The sensitivity and specificity of ESR and hsCRP for active infection were estimated with 95% confidence intervals.

Results

Figure 1 depicts the study schematic. In total, 1089 unique inpatients with a recorded diagnostic or procedure code for SLE were identified between 28 August 2001 and 27 April 2008. Of these patients, 338 had both an ESR and an hsCRP level documented. Patients in whom the lupus diagnosis was not confirmed (198 patients) and those admitted for a reason other than infection or a lupus flare (51 patients) were excluded. Four other patients were excluded because of incomplete data. In all, 85 eligible patients were included in the study. Of the eligible patients, 54 were found to have a lupus flare, 22 had active infection, and eight had both an active infection and a lupus flare. ‘Lupus flare’, as diagnosed by the treating physician, was based on findings of nephritis, malar or photosensitivity rash, pleuritis, pericarditis, hemolytic anemia, thrombocytopenia, pancytopenia, digital vasculitis, central nervous system (CNS) vasculitis, pancreatitis, or arthritis. Many patients had more than one finding. Of the patients with active infection, 12 had a positive culture recorded. Eleven patients had a repeat hospitalization; of these, eight patients had two hospitalizations and three had three hospitalizations. Only the first hospitalization for each patient was included in the analysis.

Subject characteristics are presented in Table 1. The majority of the patients (89%) were female and the mean age was 37 years. The majority of the patients were Caucasian or African American, together comprising 75% of the patients. Some 56% of the patients were on prednisone. Using the Spearman correlation test and Wilcoxon rank sum and signed rank tests, we found only age to be significantly associated with hsCRP levels. The lungs were the most common organ involved during a lupus flare, followed by the kidneys.

Table 1

Subject characteristics

Age (years) (mean ± SD)	37.46 ± 14
Sex
Male (%)	9 (11)
Female (%)	76 (89)
Race
Caucasian (%)	36 (42)
African American (%)	29 (34)
Hispanic (%)	13 (15)
Asian (%)	7 (8)
Creatinine (mean ± SD)	1.779 ± 2.42
Hemoglobin (mean ± SD)	10.34 ± 1.91
Organ involvement (%)
Kidney (%)	16 (19)
Lungs (%)	20 (24)
Heart (%)	9 (11)
Skin (%)	5 (6)
System (%)	6 (7)
Joints (%)	9 (11)
Blood (%)	7 (8)
Nervous System (%)	6 (7)
Vessel (%)	3 (4)
Gastrointestinal (%)	4 (5)
Muscle (%)	1 (1)
Medication use
Prednisone (%)	48 (56)
HCQ (%)	19 (22)
MMF (%)	7 (8)
Azathioprine (%)	9 (11)
Methotrexate (%)	2 (2)
Biological Agent (%)	7 (8)
Other DMARD (%)	2 (2)
Statins (%)	7 (8)
Hormones (%)	4 (5)

SD, standard deviation; HCQ, hydroxychloroquine; MMF, Mycophenolate Mofetil; DMARD, disease-modifying anti-rheumatic drug; Biological Agent includes Infliximab, Adalimumab, Etanercept, Rituximab; Hormones include birth controls pills, hormone replacement therapy.

Mean and median values of the documented ESR and hsCRP levels for patients with active lupus flare, infection, or both are summarized in Table 2 and Figure 2. While there was no significant difference in the ESR levels in the three groups (Kruskal–Wallis test for non-normal distribution, p = 0.7284), hsCRP levels were significantly lower in the active SLE group than in the infection group (Wilcoxon rank sum test, p = 0.0003). Seven patients in the active SLE group had significantly higher hsCRP levels than others (outliers; clear circles on the graph, Figure 2). Of these, two patients had pleuritis, one had pneumonitis, two had pericarditis, one had nephritis and another had myositis.

Table 2

Relationship between hsCRP/ESR levels, lupus flare, active infection or both

Variable	Group	N	Mean	SE Mean	SD	Median	Range
ESR	SLE flare	55	64.4	5	37.3	60	2–145
	Active infection	22	68.6	8.6	40.3	73	2–127
	Both	8	55.6	16.5	46.7	44.5	7–114
hsCRP	SLE flare	55	2.7	0.5	3.5	1.2	0.01–15.9
	Active infection	22	9.52	2	9.3	6.9	0.04–34.6
	Both	8	3.7	1.2	3.3	3.3	0.01–9.1

N, number of patients; SE, standard error; SD, standard deviation.

Using a logistic regression model, only hsCRP was found to be a significant predictor of infection (p = 0.000106, area under the curve 0.75; Figure 3). Of the confounding factors (age, sex, ethnicity, creatinine, hemoglobin, medications), only age was found to be significantly correlated with hsCRP levels. We used a logistic regression model to calculate the specificity and sensitivity of hsCRP level as a predictor of infection in SLE. Employing this model, an hsCRP level of above 5 mg/dl is associated with infection with a specificity of 80% and a sensitivity of 59%. Using an hsCRP cut-off of 6 mg/dl, infection was present with a specificity and sensitivity of 84% and 55%, respectively (Figure 3).

Patients with culture-proven infection had a significantly higher ESR level than those without (Welch two-sample t-test; p = 0.03). Although mean hsCRP level was higher in patients with culture-proven infection than in patients without, this difference was not significant (p = 0.52) (Figure 4).

The association between ESR and hsCRP levels and organ system involvement in patients with SLE flare is summarized in Table 3. Some patients had more than one organ involved. hsCRP level was significantly higher in patients with pulmonary involvement than without, and lower in patients with constitutional symptoms. When we adjusted for confounders using multivariable logistic regression the results did not change.

Table 3

Subdivision of patients with SLE flare by organ system involved (some patients may have more than one organ involved)

Organ involved	Number of patients (%)	Median CRP without organ involved	Median CRP with organ involved	CRP (p-value)	Median ESR without organ involved	Median ESR with organ involved	ESR (p-value)
Kidney	16 (19)	1.57	0.66	0.18	60	82	0.19
Lungs	20 (24)	0.69	3.34	0.01*	60	64.5	0.89
Heart	9 (11)	1.195	2.2	0.32	60	92	0.49
Skin	5 (6)	1.25	0.23	0.15	60	94	0.89
System	6 (7)	1.485	0.08	0.00*	60	65.5	0.84
Joints	9 (11)	1.25	1.05	0.53	60	60	0.43
Blood	7 (8)	1.48	0.54	0.14	60	59	0.94
NS	6 (7)	1.25	0.53	0.56	64	40	0.26
Vessel	3 (4)	1.365	0.185	0.13	61	15.5	0.05
GI	4 (5)	1.115	1.41	0.71	63	48	0.30
Muscle	1 (1)	1.23	8.66	0.19	60	42	0.60

CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; GI, gastrointestinal, NS, nervous system.

p-value <0.05.

Discussion

In 1930, Tillett and Francis observed that serum from patients infected with pneumococcal pneumonia precipitated the ‘C Substance’, a non-protein carbohydrate purified from the pneumococcus species. After patients recovered from the infection, their serum lost its ability to precipitate the C Substance. This reaction to C Substance was also found in the sera of patients with rheumatic fever, bacterial endocarditis, and lung abscesses.¹⁷ These investigators used the term ‘C-reactive protein’ to refer to the then elusive plasma protein. Subsequently, CRP was isolated and extensively characterized.¹⁸ CRP is produced by the liver and adipocytes in response to various acute and chronic inflammatory processes, and is referred to as an ‘acute-phase protein’. Its synthesis in hepatocytes is stimulated by a rise in IL-6, among other cytokines, and it binds to polysaccharides of many bacteria, fungi, and certain parasites. CRP can activate the complement system and may have a role in the clearance of apoptotic cells.^19,20

hsCRP in SLE

Serum levels of CRP usually rise in parallel with disease activity in inflammatory status such as RA; however, its behavior in SLE has been surprising and subject to controversy. Several older studies investigating the role of CRP in patients with SLE, using conventional methods of CRP measurements, concluded that while CRP levels rose significantly in SLE patients with infection, this rise was only modest at best in SLE patients with active disease without evidence of infection.^2,6 ^– ¹³ Some suggested a CRP cut-off of 5 mg/dl²¹ or 6 mg/dl,^8,9,12 above which presence of infection was highly suspected. Controversy has surrounded this issue, however, as other investigators found an elevation of serum CRP in active SLE even in absence of infection.^16,22 Although three recent studies^20,23,24 have inspected the association of hsCRP levels and organ-specific lupus activity, patients with infection were excluded from these analyses; therefore it is not known how the elevated hsCRP in these patients would compare with the level in infected patients. In addition, although Lee et al.²⁴ reported significantly higher median hsCRP levels in SLE patients with organ damage than in those without, the median hsCRP in both groups was less than 0.5 mg/dl. This is in contrast to the much higher hsCRP levels found in SLE patients with infection.

Our findings revealed that hsCRP, like conventional CRP, rises to significantly higher levels in lupus patients with infection than in those with active disease. Unlike some of the older studies, however, the hsCRP level in our patients with active SLE was not undetectable. We believe that this is explained by the fact that high-sensitivity methods detect much lower levels of rise in CRP in active SLE that would have been missed by less sensitive methods. Although we did not include patients without active SLE or infection for comparison, the mean hsCRP level of 2.66 mg/dl in the active SLE group is clearly above the cut-off of <0.8 mg/dl for healthy individuals, and indicative of some level of activity in these patients. In a recent study of 53 lupus patients, Jackish et al. found an ESR/CRP ratio to be useful in distinguishing between the two conditions.²⁵ ESR/CRP ratios greater than 15 were significantly correlated with lupus flares, while ratios less than 2 were correlated with infections. Using the ESR/hsCRP ratio in our study, we found that while the difference between the two groups was significant using the Kruskal test (p-value 0.0012) and the Wilcoxon rank test (p-value 0.000037), the significance was lost when we used logistic regression (p-value 0.17) (data not shown).

We observed that an hsCRP level greater than 5 mg/dl was associated with active infection with 80% specificity. When an hsCRP level of 6 mg/dl was used as a threshold, the specificity rose to 84%. We chose these hsCRP thresholds as suggested by previous studies.^8,9,12,21 Based on our findings, in a patient with an hsCRP level greater than 6 mg/dl, infection is the most likely cause of the elevated hsCRP.

Reasons for a blunted hsCRP in SLE subjects

Several theories have been proposed to explain the blunted hsCRP response in active SLE, compared with its robust rise in other inflammatory conditions. The low hsCRP in SLE patients can be explained by either increased clearance or decreased production of this protein. Although autoantibodies to CRP in SLE patients have been discovered,²⁶ ^– ²⁸ supporting its increased clearance, the plasma clearance rate of CRP is the same in patients with active lupus and normal individuals, making this hypothesis less likely.²⁹

The decreased production of CRP in SLE patients may be due to individual genetic differences in the capacity to respond to certain stimuli.⁸ Since CRP binds to apoptotic cells, interacts with complement and phagocytic cells, and may have a role in apoptotic cell clearance, it has been proposed that a decreased CRP production in these individuals may play a role in SLE disease predisposition and pathogenesis.⁸ ^– ^10,23,30 This hypothesis may not explain the full picture, however, as it is known that even those SLE patients that do not mount a significant hsCRP response to disease flare will have a robust rise in hsCRP in response to infection.

It appears that CRP is not the only protein that is underexpressed in SLE patients. Serum levels of serum amyloid A³¹ and fibrinogen³² are also low in SLE patients compared with RA patients. The genes for pentraxins CRP and serum amyloid A both map to a segment of chromosome 1. It is conceivable that genetic polymorphisms cause their reduced production, leading to susceptibility to SLE.³³ Since production of all three proteins is induced by IL-6, it also seems logical to conclude that either a low level of IL-6 or a defect in response to IL-6 may be the cause of reduced production of these proteins.

IL-6 and CRP

It is thought that IL-6 is the main cytokine responsible for CRP induction; therefore, a low level of IL-6 production in lupus patients could explain the low CRP levels. In fact, a low IL-6 response in SLE patients has been reported.³⁴ ^– ³⁶ However, other studies have shown a strong association between IL-6 levels and disease activity in both SLE and RA patients.³⁷ ^– ⁴¹ In these studies, while the elevated IL-6 level showed a positive correlation with CRP levels in patients with RA, it did not correlate well with CRP levels in patients with SLE.^32,39,40,42 Interestingly, a similar dissociation between IL-6 and CRP levels was found in patients with Sjögren’s disease compared with controls.⁴⁰

The deficiency of CRP production in active SLE appears to be downstream to IL-6. Since these patients are capable of producing a significant CRP response to infection, then a genetic defect in hepatocytes, antibodies to IL-6 or deficiency of other cytokines cannot be the sole explanation. Patients on or off glucocorticoids and cytotoxic medications show a similar dissociation between IL-6 and CRP;³² therefore an inhibitory effect of medications can be discounted. Taken all together, these findings support a process in active SLE itself that interrupts the relationship between an elevated IL-6 and an expected rise in CRP in these patients. It is conceivable that, as proposed by Gabay et al.,³² the IL-6 receptors on hepatocytes are downregulated in active SLE, causing the blunted response. It has been hypothesized that an increase in the expression of type I IFN in SLE may explain the disparity between IL-6 and CRP in these patients.⁴³ In this model, type I IFN poses an in inhibitory effect on two proteins that are critical for induction of the CRP gene by IL-6, namely enhancer binding protein c/EBP and STAT-3.⁴³

CRP and organ involvement in SLE

It has been suggested that an elevated CRP can occur in SLE patients in the presence of serositis,^12,24,44 polyarthritis,^45,46 and nephritis.⁴⁷ Since the widespread introduction of hsCRP methodologies, investigators have also reported a relationship between elevation of this marker and specific organ involvement in SLE. Lee and coworkers²⁴ found significantly higher hsCRP values in SLE patients with myocarditis, cardiac murmur, interstitial pulmonary fibrosis, pulmonary hypertension, gastrointestinal manifestations, and anemia than in those without. In a study of 588 SLE patients from the LUMINA cohort, Bertoli et al.²³ reported a significant association between hsCRP and the constitutional, eye, pulmonary, gastrointestinal, neuromotor, and laboratory domains of the Systemic Lupus Activity Measure-Revised (SLAM-R). Barnes et al.²⁰ found hsCRP levels to be significantly higher in lupus patients than in controls, but hsCRP was not associated with disease activity assessed with the SLEDAI.

We found significantly higher hsCRP levels in patients with pulmonary disease. While the majority of the active SLE group’s hsCRP level ranged from 0.1–5.5 mg/dl, seven patients had significantly higher hsCRP levels (8.8–15.9 mg/dl). These patients had pericarditis, pleuritis, pneumonitis, nephritis and myositis. It has been suggested that the large inflammatory cell mass and vasculitic component of the pleural/pericardial lesion may be responsible for the elevated hsCRP in this group of patients.⁴⁴ Swaak et al.³⁴ found a positive association between IL-6 and CRP levels only in those lupus patients who had a disease exacerbation accompanied by serositis. This finding suggests that in these patients, the association between IL-6 and CRP is not interrupted as it seems to be in other organ involvement. However, not all our patients with pleuritis and pericarditis exhibited this trend; thus it is unclear which patients with pleural involvement will show the higher hsCRP. Unfortunately, information on the patients’ IL-6 level was not available in our patients.

Conditions affecting ESR and CRP

ESR, a measure of the distance that erythrocytes have fallen after 1 h in a vertical column of anticoagulated blood, is an indirect measure of inflammation and is influenced by a variety of factors.^48,49 Conditions such as gender, age, renal disease, anemia, heart failure, and obesity, among others, can cause wide fluctuations in ESR levels.^48,50 ^– ⁵² While as a direct measurement of an acute-phase plasma protein CRP may be a more reliable measure of inflammation,⁵³ its levels are also influenced by a variety of factors. Ethnicity,⁵⁴ gender,⁵⁵ body mass index (BMI),⁵⁶ oral contraceptives,⁵⁷ and hormone replacement therapy⁵⁸ as well as chronic renal failure^16,47 have been associated with elevated CRP levels, while statins,^1,59 anti-malarials²⁰ and corticosteroids have been thought to reduce its levels.⁶⁰

We noted a statistically significant correlation only between hsCRP levels and the patients’ age. We observed no statistically significant correlation between hsCRP levels and gender, ethnicity, hemoglobin level, creatinine, or any medications in our lupus patients. As data on the patients’ BMI or smoking habits was not available, we were not able to determine the effect of these potential confounders on hsCRP levels.

This study had several limitations. Due to the retrospective nature of the study, we had to rely on the interpretation of the treating physician to determine whether or not a patient had a lupus flare. Reanalyzing our data without the outliers may also be perceived as a limitation, as the decision to do so was made after the initial review of data and the realization that the majority of these patients in fact had serositis. We decided on this separate analysis based on the fact that prior studies had found an association between serositis and an elevated hsCRP. In addition, although BMI has been shown to impact levels of both ESR and hsCRP, since we were not able to find the height and weight simultaneously on the electronic medical records, we were not able to assess the role of obesity as a confounder. Finally, since our data were limited to hospitalized patients, it is unclear how our findings can be applied to outpatients with lupus. We did not compare the positive predictive value of hsCRP in SLE patients to that of conventional CRP, and as such can not suggest superiority of hsCRP to conventional CRP for these or similar analyses.

Conclusions

Our observations revealed that hsCRP was significantly lower in SLE patients with active disease than in those with either documented or suspected infections. We propose that elevated hsCRP levels can be used as a predictor of active infection with a specificity of 80% at levels greater than 5 mg/dl and 84% at levels greater than 6 mg/dl. A serum hsCRP level greater than 6 mg/dl in a patient with SLE is a strong predictor of active infection. Although IL-6 is thought to be the main cytokine responsible for CRP production, coupled with our observations that there is not a robust CRP response in active SLE, further investigations are awaited especially since IL-6 could be a target for therapeutic purposes. These studies might clarify some important relationships that otherwise remain unexplained.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare that they have no conflicts of interest.

References

Ganapathi

Rzewnicki

Samols

Jiang

Kushner

. Effect of combinations of cytokines and hormones on synthesis of serum amyloid A and C-reactive protein in Hep 3B cells. J Immunol 1991; 147: 1261–1265.

Pereira Da Silva

Elkon

Hughes

Dyck

Pepys

. C-reactive protein levels in systemic lupus erythematosus: a classification criterion? Arthritis Rheum 1980; 23: 770–771.

Coste

Spira

Clerc

Paolaggi

. Prediction of articular destruction in rheumatoid arthritis: disease activity markers revisited. J Rheumatol 1997; 24: 28–34.

Wolfe

. Comparative usefulness of C-reactive protein and erythrocyte sedimentation rate in patients with rheumatoid arthritis. J Rheumatol 1997; 24: 1477–1485.

Cantini

Salvarani

Olivieri

. Erythrocyte sedimentation rate and C-reactive protein in the diagnosis of polymyalgia rheumatica. Ann Intern Med 1998; 128: 873–874.

Hill

. C-reactive protein in the chronic rheumatic diseases. Lancet 1951; 2: 807–811.

Honig

Gorevic

Weissmann

. C-reactive protein in systemic lupus erythematosus. Arthritis Rheum 1977; 20: 1065–1070.

Becker

Waldburger

Hughes

Pepys

. Value of serum C-reactive protein measurement in the investigation of fever in systemic lupus erythematosus. Ann Rheum Dis 1980; 39: 50–52.

Pepys

Hirschfield

Tennent

. Targeting C-reactive protein for the treatment of cardiovascular disease. Nature 2006; 440: 1217–1221.

10.

Hind

Feng

Pepys

. Serum C-reactive protein measurement in the detection of intercurrent infection in Oriental patients with systemic lupus erythematosus. Ann Rheum Dis 1985; 44: 260–261.

11.

Bravo

Alarcon-Segovia

. C-reactive protein in the differential diagnosis between infection and disease reactivation in SLE. J Rheumatol 1981; 8: 291–294.

12.

ter Borg

Horst

Limburg

van Rijswijk

Kallenberg

. C-reactive protein levels during disease exacerbations and infections in systemic lupus erythematosus: a prospective longitudinal study. J Rheumatol 1990; 17: 1642–1648.

13.

Bertouch

Roberts-Thompson

Feng

Bradley

. C-reactive protein and serological indices of disease activity in systemic lupus erythematosus. Ann Rheum Dis 1983; 42: 655–658.

14.

Koenig

Sund

Frohlich

. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999; 99: 237–242.

15.

Danesh

Wheeler

Hirschfield

. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med 2004; 350: 1387–1397.

16.

Williams

Jr Harmon

Burlingame

Du Clos

. Studies of serum C-reactive protein in systemic lupus erythematosus. J Rheumatol 2005; 32: 454–461.

17.

Tillett

Francis

. Serological reactions in pneumonia with a non-protein somatic fraction of pneumococcus. J Exp Med 1930; 52: 561–571.

18.

Lotze

Thompson

(eds), Roles of acute phase protein in inflammation, immune defense and disease. In: Measuring immunity: basic biology and clinical assessment. Amsterdam: Elsevier Academic Press; 2005. 135–135.

19.

Ledue

Weiner

Sipe

Poulin

Collins

Rifai

. Analytical evaluation of particle-enhanced immunonephelometric assays for C-reactive protein, serum amyloid A and mannose-binding protein in human serum. Ann Clin Biochem 1998; 35: 745–753.

20.

Barnes

Narain

Naranjo

. High sensitivity C-reactive protein in systemic lupus erythematosus: relation to disease activity, clinical presentation and implications for cardiovascular risk. Lupus 2005; 14: 576–582.

21.

Suh

Jeong

Park

. Risk factors for infection and role of C-reactive protein in Korean patients with systemic lupus erythematosus. Clin Exp Rheumatol 2001; 19: 191–194.

22.

al-Mekaimi

Malaviya

Serebour

. Serological characteristics of systemic lupus erythematosus from a hospital-based rheumatology clinic in Kuwait. Lupus 1997; 6: 668–674.

23.

Bertoli

Vila

Reveille

Alarcon

. Systemic lupus erythematosus in a multiethnic US cohort (LUMINA): LXI. Value of C-reactive protein as a marker of disease activity and damage. J Rheumatol 2008; 35: 2355–2358.

24.

Lee

Singh

Link

Petri

. High-sensitivity C-reactive protein as an associate of clinical subsets and organ damage in systemic lupus erythematosus. Semin Arthritis Rheum 2008; 38: 41–54.

25.

Jackish

Somers

McCune

. Comparison of ESR (erythrocyte sedimentation rate) and CRP (C-reactive protein) in lupus patients presenting with fever. Abstract. American College of Rheumatology Annual Scientific Meeting. Washington DC; 2006. L22–L22.

26.

Bell

Faust

Schmid

Meurer

. Autoantibodies to C-reactive protein (CRP) and other acute-phase proteins in systemic autoimmune diseases. Clin Exp Immunol 1998; 113: 327–332.

27.

Sjowall

Bengtsson

Sturfelt

Skogh

. Serum levels of autoantibodies against monomeric C-reactive protein are correlated with disease activity in systemic lupus erythematosus. Arthritis Res Ther 2004; 6: R87–R94.

28.

Sjowall

Wettero

. Pathogenic implications for autoantibodies against C-reactive protein and other acute phase proteins. Clin Chim Acta 2007; 378: 13–23.

29.

Vigushin

Pepys

Hawkins

. Metabolic and scintigraphic studies of radioiodinated human C-reactive protein in health and disease. J Clin Invest 1993; 91: 1351–1357.

30.

Marnell

Mold

Du Clos

. C-reactive protein: ligands, receptors and role in inflammation. Clin Immunol 2005; 117: 104–111.

31.

Nijsten

de Groot

ten Duis

Klasen

Hack

Aarden

. Serum levels of interleukin-6 and acute phase responses. Lancet 1987; 2: 921–921.

32.

Gabay

Roux-Lombard

de Moerloose

Dayer

Vischer

Guerne

. Absence of correlation between interleukin 6 and C-reactive protein blood levels in systemic lupus erythematosus compared with rheumatoid arthritis. J Rheumatol 1993; 20: 815–821.

33.

Russell

Cunninghame Graham

Shepherd

. Polymorphism at the C-reactive protein locus influences gene expression and predisposes to systemic lupus erythematosus. Hum Mol Genet 2004; 13: 137–147.

34.

Swaak

van Rooyen

Aarden

. Interleukin-6 (IL-6) and acute phase proteins in the disease course of patients with systemic lupus erythematosus. Rheumatol Int 1989; 8: 263–268.

35.

Swaak

van den Brink

Aarden

. Cytokine production (IL-6 and TNF alpha) in whole blood cell cultures of patients with systemic lupus erythematosus. Scand J Rheumatol 1996; 25: 233–238.

36.

Peterson

Robertson

Emlen

. Serum and urinary interleukin-6 in systemic lupus erythematosus. Lupus 1996; 5: 571–575.

37.

Hirohata

Miyamoto

. Elevated levels of interleukin-6 in cerebrospinal fluid from patients with systemic lupus erythematosus and central nervous system involvement. Arthritis Rheum 1990; 33: 644–649.

38.

Linker-Israeli

Deans

Wallace

Prehn

Ozeri-Chen

Klinenberg

. Elevated levels of endogenous IL-6 in systemic lupus erythematosus. A putative role in pathogenesis. J Immunol 1991; 147: 117–123.

39.

Lacki

Samborski

Mackiewicz

. Interleukin-10 and interleukin-6 in lupus erythematosus and rheumatoid arthritis, correlations with acute phase proteins. Clin Rheumatol 1997; 16: 275–278.

40.

Stuart

Littlewood

Maddison

Hall

. Elevated serum interleukin-6 levels associated with active disease in systemic connective tissue disorders. Clin Exp Rheumatol 1995; 13: 17–22.

41.

Chun

Chung

Kim

. Cytokine IL-6 and IL-10 as biomarkers in systemic lupus erythematosus. J Clin Immunol 2007; 27: 461–466.

42.

Meijer

Huysen

Smeenk

Swaak

. Profiles of cytokines (TNF alpha and IL-6) and acute phase proteins (CRP and alpha 1AG) related to the disease course in patients with systemic lupus erythematosus. Lupus 1993; 2: 359–365.

43.

Gaitonde

Samols

Kushner

. C-reactive protein and systemic lupus erythematosus. Arthritis Rheum 2008; 59: 1814–1820.

44.

Mochizuki

Aotsuka

Satoh

. Clinical and laboratory features of lupus patients with complicating pulmonary disease. Respir Med 1999; 93: 95–101.

45.

Moutsopoulos

Mavridis

Acritidis

Avgerinos

. High C-reactive protein response in lupus polyarthritis. Clin Exp Rheumatol 1983; 1: 53–55.

46.

Spronk

ter Borg

Kallenberg

. Patients with systemic lupus erythematosus and Jaccoud’s arthropathy: a clinical subset with an increased C reactive protein response? Ann Rheum Dis 1992; 51: 358–361.

47.

Zuniga

Markowitz

Arkachaisri

Imperatore

D’Agati

Salmon

. Identification of IgG subclasses and C-reactive protein in lupus nephritis: the relationship between the composition of immune deposits and FCgamma receptor type IIA alleles. Arthritis Rheum 2003; 48: 460–470.

48.

Brigden

. Clinical utility of the erythrocyte sedimentation rate. Am Fam Physician 1999; 60: 1443–1450.

49.

Costenbader

Chibnik

Schur

. Discordance between erythrocyte sedimentation rate and C-reactive protein measurements: clinical significance. Clin Exp Rheumatol 2007; 25: 746–749.

50.

Bedell

Bush

. Erythrocyte sedimentation rate. From folklore to facts. Am J Med 1985; 78: 1001–1009.

51.

Ballou

. Laboratory evaluation of inflammation. In: Kelly’s textbook of rheumatology. Philadelphia, PA: Saunders Elsevier; 2005. 720–727.

52.

Brigden

. The erythrocyte sedimentation rate. Still a helpful test when used judiciously. Postgrad Med 1998; 103: 257–262. 72–74).

53.

Dilber

Cakir

Kalyoncu

Okten

. C-reactive protein: a sensitive marker in the management of treatment response in parapneumonic empyema of children. Turk J Pediatr 2003; 45: 311–314.

54.

Khera

McGuire

Murphy

. Race and gender differences in C-reactive protein levels. J Am Coll Cardiol 2005; 46: 464–469.

55.

Albert

Glynn

Buring

Ridker

. C-reactive protein levels among women of various ethnic groups living in the United States (from the Women’s Health Study). Am J Cardiol 2004; 93: 1238–1242.

56.

Visser

Bouter

McQuillan

Wener

Harris

. Elevated C-reactive protein levels in overweight and obese adults. JAMA 1999; 282: 2131–2135.

57.

van Rooijen

Hansson

Frostegard

Silveira

Hamsten

Bremme

. Treatment with combined oral contraceptives induces a rise in serum C-reactive protein in the absence of a general inflammatory response. J Thromb Haemost 2006; 4: 77–82.

58.

Cushman

Legault

Barrett-Connor

. Effect of postmenopausal hormones on inflammation-sensitive proteins: the Postmenopausal Estrogen/Progestin Interventions (PEPI) Study. Circulation 1999; 100: 717–722.

59.

Ridker

Cannon

Morrow

. C-reactive protein levels and outcomes after statin therapy. N Engl J Med 2005; 352: 20–28.

60.

Zonana-Nacach

Barr

Magder

Petri

. Damage in systemic lupus erythematosus and its association with corticosteroids. Arthritis Rheum 2000; 43: 1801–1808.