Serum CRP Levels in Pre-Dialysis Patients

Abstract

Background. An elevated serum C-reactive protein (CRP) is strongly associated with morbidity and mortality in dialysis patients. However, the significance of high CRP levels in pre-dialysis patients has not been studied extensively. The aim of this study was to determine the prevalence of elevated serum CRP in pre-dialysis patients and to analyze its correlation with renal function and other inflammatory and nutritional factors. Methods. In a cross-sectional study, 100 pre-dialysis patients who had been visited in two outpatient nephrology clinics from 2005 until 2006 and had the serum creatinine ≥ 1.5 mg/dL for at least three months were studied. Demographic characteristics, medications, GFR, hemoglobin, as well as inflammatory and nutritional parameters (CRP, Albumin, Fibrinogen, Transferin, Ferritin, TG, Chol, LDL, and HDL) were measured and compared between the patients in regard to the CRP level. Results. The mean of serum CRP level was 5.7 ± 5.1mg/L; elevated level were reported in 17 patients (17%). Serum CRP levels was significantly correlated with GFR, albumin, fibrinogen, transferring, and ferritin. Conclusion. Similar to the dialysis population, we found that serum CRP was elevated in pre-dialysis patients. In addition, a positive correlation between serum CRP levels and several inflammatory factors was found. CRP serum level was also negatively correlated with GFR, the indicator of renal function.

Keywords:

• dialysis
• C-reactive protein
• GFR
• serum albumin

INTRODUCTION

Despite the remarkable advances in the field of dialysis within the last 20 years, the mortality rate in patients with end stage renal disease (ESRD) is quite high. Cardiovascular diseases, with a mortality rate of 9%, are the major cause of death in this group; this rate is 10–20 times higher than the normal population even after age, gender, race, and diabetes mellitus adjustments. Such a high rate indicates the presence of an accelerated atherogenesis process.[1],[2]

In addition to traditional risk factors of arthrosclerosis, uremia and dialysis-related factors may also release the pre-inflammatory cytokines and disturb the endothelial performance. It may also produce an acute or chronic systemic inflammatory response (increase in the C-reactive protein [CRP] level and other proathrotrombic factors), consequently accelerating the arthrosclerosis process. Therefore, inflammation has a major role in arthrosclerosis in ESRD patients.[3],[4]

C-reactive protein has the most important role in the inflammatory response and is the most common index for diagnosing inflammation.[5],[6] An elevated serum CRP is reported in 20–65% of ESRD patients (pre-dialysis and those under hemodialysis and peritoneal dialysis). This increase in serum CRP and other acute-phase proteins are caused by underlying factors that lead to acute phase responses and the activation of the inflammatory cascade.[7],[8] However, a reduction in renal clearance of pre-inflammatory cytokines in addition to diseases and their accompanying complications, like cardiac failure and advanced glycation end product (AGEs) accumulation, as well as dialysis-related factors can cause inflammation and increase the serum CRP levels in ESRD patients.[1],[9],[10]

Several studies have proved the relationship between an elevated serum CRP levels and renal function (GFR reduction), arthrosclerosis, malnutrition, low serum albumin, anemia, low hemoglobin resistant to erythropoietin, and frequent hospitalizations, as well as general morbidity and mortality due to cardiovascular diseases in patients undergoing hemodialysis or peritoneal dialysis.[11],[12] Few studies have been carried out to study the correlation between the inflammation and the serum CRP levels in pre-dialysis patients (patients with decreased GFR who do not require dialysis or kidney transplant) and its relationship with renal function.

The population of pre-dialysis patients consists of a higher number of patients compared to dialysis group; as they are different from each other in various aspects, the potential inflammation-causing factors in pre-dialysis patients may be different from those of dialysis patients. Moreover, because the renal function of pre-dialysis patients has been observed in various stages, their inflammatory conditions may also be different. Thus, in the present study, the serum CRP levels and its correlation with renal function and other inflammatory and nutritional factors were studied in patients with chronic renal failure referred to two outpatient nephrology clinics.

MATERIALS AND METHODS

Pre-dialysis patients with a creatinine levels higher than 1.5 mg/dL for at least three months referred to two general hospitals in Tehran, Iran, in 2005 were enrolled in this study. They were all older than 70 years old. Those with a positive history of MI and CCU admission (for more than three days) within a year prior to this study; those with a positive history of infectious diseases within a six-month period; those in the active phase of vasculitis, malignancy, or amyloidosis; and those administrating nitrates were excluded from the study.

The study was conducted according to the Helsinki principles. Informed consent was taken from all patients. The nephrologists examined the patients thoroughly, and demographic data, past medical history (cardiovascular diseases), and administered medications were recorded. Laboratory findings, including albumin, creatinine, BUN, CRP, CBC-Diff, lipid profile, ferritin, transferrin, and fibrinogen, were also recorded; all samples were sent to Tehran Noor Pathobiological Laboratory. GFR was calculated in each patient using Cockcroft-Gault formula. Serum albumin and CRP were measured using Bromocrosol Green and photometry, respectively.

According to their CRP serum levels, patients were divided into two groups:

• The CRP negative group: patients with CRP <10 mg/L

• The CRP positive group: patients with serum CRP ≥10 mg/L

The data were analyzed using SPSS 10 software. Mann-Whitney U (variables without normal distribution) and T-test (variables with normal distribution) were used to compare the quantitative variables between the above-mentioned groups; chi-square test was also used for comparing qualitative variables. Linear regression was used to study the correlation between the two quantitative variables; amounts with p < 0.05 were considered statistically significant.

RESULTS

One hundred pre-dialysis patients with chronic renal failure were studied; 55 of which were male. The mean age of the patients was 55.8 ± 13.32, years, ranging between 21 and 70 years. Hypertension (65%), diabetes (33%), and unknown etiology (23%) were the most common causes of CRF in our cases. Table 1 outlines the underlying diseases for renal failure in these cases.

Table 1 The underlying disease for renal failure in these cases

		CI 95%
Renal diseaseetiology	Number of patients	Lower	Upper
DM	8	2.68	13.32
HTN	40	30.39	49.61
DM&HTN	25	16.51	33.49
GN	2	0	4.75
PKD	1	0	2.96
Idiopathic	23	14.75	31.25
TB	1	0	2.96
Underlying factors
Heart disease	22	13.88	30.12
Drugs used
ASA	18	10.47	25.53
Statins	14	7.19	20.81
ACE Inhibitors	41	31.36	50.64
Vitamin E	29	20.1	37.9
Sex
Male	55	45.24	64.76
Female	45	35.24	54.76

Abbreviations: GN = glomerulonephritis, PKD = polycystic kidney disease

The demographic data and laboratory findings are described in Table 2. Table 3 demonstrates the laboratory findings in the two groups. There was no significant difference between the demographic data of the two groups. The mean serum CRP level of the patients was 5.8 ± 5.1 mg/L (0.5–37). Seventeen patients were diagnosed to be CRP positive. The findings of the present study revealed no statistically meaningful difference between the amounts of triglyceride, cholesterol, LDL, and HDL levels in either group (p > 0.05).

Table 2 The demographic data and laboratory findings of the enrolled patients

Variables	Mean	Median	Standard deviation	Minimum	Maximum
Age	55.76	60	13.32	21	70
GFR (mL/min)	32.04	28.85	16.34	5	74.7
Albumin (gr/dL)	4.25	4.3	0.65	2.1	5.4
Fibrinogen (mg/dL)	360.05	341	148.60	72	770
Transferin (mg/dL)	262.46	270	94.51	70	700
Triglycerides (mg/dL)	180.53	160	97.95	60	521
Cholesterol (mg/dL)	194.66	188.5	52.42	76	348
HDL (mg/dL)	50.91	45.5	25.20	16	188
LDL (mg/dL)	92.52	89	40.06	15	227
Hemoglobin (g/dL)	11.96	12	2.39	5.4	18.1
Ferritin (ng/mL)	238.95	133	278.33	17	2000
CRP (mg/L)	5.77	4.5	5.16	0.5	37

Table 3 The laboratory findings in the two groups (in regard with CRP levels)

	CRP 10 < mg/L		CRP ≥10 mg/L				CI 95% of mean difference
	N = 83		N = 17
Variables	Mean	SD	Mean	SD	p	Mean difference	Lower	Upper
Age	55.36	13.42	57.71	13.00	0.511	−2.34	−9.40	4.71
GFR (mL/min)	33.61	16.33	24.32	14.48	0.032	9.29	0.82	17.77
Albumin (gr/dL)	4.30	0.62	4.00	0.75	0.082	0.30	−0.04	0.64
Fibrinogen (mg/dL)	345.61	134.63	430.53	193.32	0.100	−84.91	−161.96	−7.87
Transferin (mg/dL)	271.88	95.41	216.47	76.85	0.027	55.41	6.47	104.35
Triglycerides (mg/dL)	184.76	102.02	159.88	74.03	0.343	24.88	−26.90	76.65
Cholesterol (mg/dL)	198.20	51.62	177.35	54.43	0.136	20.85	−6.67	48.37
HDL (mg/dL)	52.19	26.77	44.65	14.42	0.263	7.55	−5.75	20.84
LDL (mg/dL)	93.41	38.68	88.18	47.30	0.626	5.23	−16.01	26.48

Table 4 shows the correlation between different laboratory findings and CRP levels. The mean GFR in the CRP-negative group was significantly lower than that in the CRP-positive group (<10 mg/L) (24.3 ± 14.4 mL/min vs. 33.6 ± 16.3 mL/min, p = 0.03); serum CRP level was significantly correlated with GFR (r  =  −0.256, p = 0.01). Serum transferrin was the other factor reported to be significantly different between the two groups; it was reported to be adversely related with the serum CRP level (r  =  −0.3, p = 0.002). It should be noted that ferritin and fibrinogen were positively correlated with CRP (r  =  0.3 and 0.2, respectively).

Table 4 The correlation between different laboratory findings and CRP level

Variables	Pearson CC	p
Age	0.18	0.074
GFR (mL/min)	−0.26	0.010
Albumin (gr/dL)	−0.20	0.046
Fibrinogen (mg/dL)	0.27	0.007
Transferin (mg/dL)	−0.31	0.002
Triglycerides (mg/dL)	−0.09	0.373
Cholesterol (mg/dL)	−0.20	0.045
HDL (mg/dL)	−0.04	0.714
LDL (mg/dL)	−0.05	0.601
Hemoglobin (g/dL)	−0.16	0.115
Ferritin (ng/mL)	0.35	0.000

This study shows the difference between the mean serum hemoglobin of the CRP-positive (≥10 mg/L) and the CRP-negative (<10 mg/L) groups were not statistically significant (11.2 ± 2.5 mg/L vs. 12.1 ± 2.3 mg/L, p = 0.2), and the serum CRP level and hemoglobin level of patients were not correlated (r  =  −0.16, p = 0.115).

The prevalence of a CRP-positive (≥10mg/L) result in patients using ASA (11.1% vs. 18.3%, p = 0.4), statins (14.3% vs. 17.4%, p = 0.7), ACEI (14.6% vs. 18.6%, p = 0.6), or vitamin E (13.8 vs. 18.3%, p = 0.5) was lower than the patients who did not use the said medications.

DISCUSSION

Activated acute phase response is shown to be prevalent in dialysis patients[13]; however, few researches have studied the increase of serum CRP in pre-dialysis patients. Various rates of increase have been shown in these studies. Panichi et al.[14] reported the increased serum CRP (>5 mg/L) in 42% of the patients (mean creatinine clearance of 36.3 ± 23.1 mL/min). Stenvinkel et al.[15] stated that 32% of their patients (mean creatinine clearance of 7 ± 1 mL/min) to have an elevated serum CRP level (≥10 mg/L). In another study, this rate (CRP > 6 mg/L) reached 35% (mean creatinine clearance of 14 ± 4 mL/min).[16] The frequency of serum CRP levels reported in the present study was lower than the previous ones. Several reasons may explain such discrepancy. The serum CRP was quantitatively measured using the nephelometry method in previous studies; thus, various figures were considered as the normal value. In the present study, the amounts less than 10 mg/L were considered as normal; this justifies the lower frequency of patients with elevated serum CRP levels. On the other hand, the target population of the previous studies included patients with a lower GFR levels and a more progressive chronic renal failure, as compared with ours. It is noteworthy that the incidence of elevated serum CRP level in Asian patients is lower than American and Europeans.[17] The very concept indicates that causes other than dialysis-related factors such as differences in lifestyle and nutritional habits justify the increased serum CRP in Asian patients.[18],[19]

The underlying mechanism of activated acute phase response in pre-dialysis patients is not clearly specified. In our study, the mean GFR in the CRP-positive group was significantly lower than the CRP-negative group. This indicates that any decrease in CRP clearance may activate the acute phase response. Indeed, the inflammatory process and the elevated serum CRP levels in patients with chronic renal failure reduce GFR.[1],[7] Ates and Panichi[14],[20] also found a negative correlation between the serum CRP and GRF levels; however, the said correlation was not found in other studies.[16]

The acute phase response is set by several pre-inflammatory cytokines (IL-6, TNF-α, INF-β, INF-γ, IL-1).[21] Any reduction in renal clearance of the cytokines is the other possible cause of elevated serum CRP levels in pre-dialysis patients. The positive correlation between creatinine clearance, cytokines, and their soluble receivers is proved in various stages of renal failure.[14],[22]

Creatinine clearance is introduced as an independent factor influencing the serum CRP level of pre-dialysis patients[20]; however, the abnormal distribution of CRP levels in these patients, like the present study, indicates renal dysfunction is not solely responsible for inflammatory responses. As a result, other factors, including infections of teeth and gum as well as Chlamydia pneumonia, often accompanied with renal failure may activate the acute phase response.

Moreover, some medications administered in CRF patients may also affect the acute phase response. Recent studies have shown that the use of ACE inhibitors in ESRD patients to be accompanied with significant reduction in serum CRP levels.[23] It is also stated that taking aspirin is correlated with serum CRP level in patients with cardiac angina.[24] Recent studies have reported that statins have significant anti-inflammatory effects and reduce the serum CRP level in patients with or without renal dysfunction.[25] Using vitamin E is also shown to be accompanied with reduction in CRP and IL-6 levels[26]; high doses of vitamin E decrease cardiovascular and myocardial infarction-related mortalities.[27] On the contrary, CRP levels were not reported to be different in our patients who used aspirin, ACE inhibitors, statins, or vitamin E. This may be due to the limitations of the present study: having no control groups and not making follow-ups, on the one hand, and enrolling patients who used different doses of the mentioned medications, on the other, may be the reasons.

Albumin and transferrin are among the negative acute phase proteins, the synthesis of which reduces during the inflammation. On the contrary, ferrtin and fibrinogen are positive acute phase proteins.[16] In our study, a statistical negative correlation was found between the serum CRP levels and the patients' negative acute phase proteins, while positive acute phase proteins were positively correlated with CRP levels. This was similar to the findings of other studies[14],[16],[20]; however, the negative correlation between serum CRP and serum transferrin was only reported in the study of Ates et al.[20]

Contrary to Ortega's study,[16] Ates et al.[20] reported a highly positive correlation between serum CRP and ferritin. Stenvinkel et al.[15] reported the correlation between serum CRP and fibrinogen levels in 109 pre-dialysis patients. In the present study, the serum CRP level was also highly correlated with fibrinogen and ferritin levels. Similar to other studies, our study revealed no statistically meaningful difference between the amounts of triglyceride, cholesterol, LDL, and HDL in either of the groups.[15],[16],[20]

Anemia is an independent risk factor for cardiovascular diseases and general mortality in patients with chronic renal failure.[28] The activated acute phase response in dialysis patients results in anemia and resistance to erythropoietin; this may be through several potential mechanisms, including suppressing erythropoiesis in the bone marrow, reducing the secretion of erythropoietin, GI bleeding, and disorders in iron metabolism.[29] A significant negative correlation between serum CRP and hemoglobin levels of pre-dialysis patients was reported in the studies of Ortega[16] and Ates.[20] However, in the present study, the serum CRP and hemoglobin levels were not correlated. It is possible that a more acute inflammation is required for any remarkable decrease in serum hemoglobin occurrence.

Similar to dialysis patients, the prevalence of elevated serum CRP seems to be high in pre-dialysis patients. Reducing inflammation plus routine measurement of CRP in these patients may improve the nutritional and cardiovascular condition and consequently slow down the renal failure process and extend the patients' lifespan.