Evaluation of Serum C-reactive Protein Level and its Related Factors in Hemodialysis Patients in Sari, Iran

MATERIALS AND METHODS

In February 2011, in a cross-sectional study from two hemodialysis centers (Imam Khomeini and Fatemeh Zahra Hospitals), 147 hemodialysis patients, aged between 20-70 years, were studied. All included patients were on chronic hemodialysis for minimum of six months and clinically stable without any obvious active inflammatory state or infectious disease. Patients with history of renal transplantation, overt cardiovascular, liver and autoimmune disease or malignancy were excluded. Also, patients who had a history of vitamin C, alcohol, oil fish and immunosuppressive drugs consumption during 2 months prior to the study were excluded. Dialysis was performed three times a week, 3.5-4 h session and the bicarbonate-based dialysate was used for all patients. Demographic and clinical characteristics of patients were obtained from interview and chart reviews. Approval from Mazandaran University of medical sciences Ethics Committee and informed written consent from patients obtained. At the beginning of the hemodialysis session, 10 mL of blood sample was drawn from venous end of Arteriovenous (AV) fistula for measuring of CRP, Cholesterol, Albumin, Phosphorous, Calcium, Hemoglobin and Hematocrit level. Serum CRP concentration was measured by Nephelometric method (Nycocard, England, Norway). Phosphorus level was measured with Pars azmoon company kit and photometric technique by BT 3000 and serum calcium was measured with ortho-cresolphthalein Complexone (o-CPC). Serum cholesterol and albumin levels were measured by standard kits. Hemoglobin and Hematocrit level was measured with cell counter.

Statistical analysis: Statistical analysis was performed with SPSS version 17 statistical software package. Data were collected and then, analyzed by chi-square, paired t test and fisher statistical tests. Statistical correlations were assessed using bivariate correlation test. The p<0.05 has been considered as statistically significant level.

RESULTS

A total of 147 hemodialysis patients were studied. The mean age of patients was 60.38±13.5 years, ranging from 25 to 70 years. The mean duration of hemodialysis was 38.82 months (range from 6 to 214 months). The mean of BMI (Body Mass Index) in patients was 24.62 (95% CI characteristics of the patients are shown in Table 1. 23.77 to 25.47). The demographic and clinical Overall, the mean CRP concentration was 15.8 mg L^-1 (95% CI: 13.7 to 17.9). Female patients had higher CRP value than men; however the difference was not statistically significant. Overall laboratory values are shown in Table 2. With considering the different cutoff point for CRP level (5, 6.2, 10 mg L^-1), 107 patients (72.8%) had CRP level>5 mg L^-1, 99 patients (67.3%) had CRP level>6.2 mg L^-1 and 77 patients (52.4%) had CRP level> 10 mg L^-1 (Table 3). CRP levels greater than 5 mg L^-1, as a cutoff point, didn’t have statistically significant correlation with any of the demographics, clinical and laboratories parameters. Relationship between CRP levels greater than 6.2, as a cutoff point, with duration of hemodialysis and phosphorus level was statistically significant. So, higher phosphorus level [in CRP<6.2= 5.5±1.7 and in CRP>6.2= 6.3±1.8 mg dL^-1 (p = 0.01)] and more time being on hemodialysis [in CRP<6.2 = 27.9 and in CRP>6.2 = 44.1 months (p=0.02)] has been associated with higher CRP levels. Also, considering CRP cutoff point above 10 mg L^-1 had a direct statistically significant correlation with age and phosphorus levels. So that the mean age of patients with CRP levels lower than 10 mg L^-1 was equal to 57.8 and in patients with higher CRP levels was equal to 62.7 years (p = 0.02).

Table 1:	Frequency distribution of patient's demographic and clinical characteristics

Table 2:	Laboratory values in hemodialysis patients (mean and 95% confidence interval)

Table 3:	Frequency distribution of patients with considering different CRP cutoff point

Table 4:	The relationship between different cutoff point of CRP with some demographic and clinical characteristic

In addition in patients with CRP levels lower and higher than 10 mg L^-1, the mean blood phosphorus level was 5.7 and 6.3 mg dL^-1, respectively (p = 0.03) (Table 4). Moreover, a significant correlation between CRP greater than 10 mg L^-1 and fixed night shift of hemodialysis was found (p = 0.04). No significant correlation between BMI, cholesterol, albumin, hemoglobin, hematocrit, calcium level and CRP was found (p>0.05).

DISCUSSION

With considering to the different cutoff point (5, 6.2, 10 mg L^-1) for CRP level in hemodialysis patients (Racki et al., 2006; Hung et al., 2008; Iseki et al., 1999), results of the present study showed that a significant proportion of hemodialysis patients had elevated level of CRP. Indeed 72.8% of patients had CRP level over the upper limit of normal human subjects (5 mg L^-1), 67.3% had CRP level>6.2 mg L^-1 and 52.4% had CRP level>10 mg L^-1. In view of the fact that hemodialysis procedure itself contributes to the inflammatory response (Yao et al., 2004), chronic inflammation may be one of the causes of increased mortality and morbidity in this population (Yilmaz, 2007), especially because of its association to atherogenesis and cardiovascular events which account for approximately 50% of the death in dialysis patients (Collins et al., 2003). Several studies have confirmed that inflammation, as reflected by elevated level of CRP, is a significant independent predictor of mortality in dialysis patients (Yeun et al., 2000; Wanner and Metzger, 2002). Korevaar et al. (2004) identified that CRP level itself and an increase in CRP level during a dialysis session are independently associated with an increased mortality risk in hemodialysis patients. In a study conducted by Racki et al. (2006), the CRP cutoff point of 6.2 mg L^-1 was predictive of increased mortality in patients with ESRD. Finding of present study showed that more than two-third of patients had CRP concentration above 6.2 mg L^-1. In another study that performed by Iseki et al. (1999) they used 10 mg L^-1 as their cutoff values in dialysis patients. With considering 10 mg L^-1 as a CRP cutoff in hemodialysis patients, 52.4% of our patients have CRP higher than 10 mg L^-1. This indicates the high incidence of inflammation in this patients. Razeghi et al. (2008) observed that in 41% of hemodialysis patients, CRP was higher than 10 mg L^-1. In a study by Hung et al. (2008) the mean of CRP was 14.3 mg L^-1 and 53% of patients had CRP level>5 mg L^-1. Increase of 1 mg L^-1 in CRP level was associated with 9% increasing in mortality risk (Korevaar et al., 2004). Iseki et al. (1999) showed a 3.5 times higher mortality risk in hemodialysis patients with CRP level greater than 5 mg L^-1 after 5 years of follow up. In all studies with different CRP cutoff points, patients with CRP values above chosen cutoff point were at increased risk of overall and cardiovascular mortality. Also, in dialysis patients high CRP level predict lower serum albumin concentration and lower serum Hemoglobin (Hb) level with poor response to EPO (Ortega et al., 2002). Teruel et al. (2005) found an inverse correlation between CRP and serum albumin and Hb level in hemodialysis patients. Also, in another study in hemodialysis patients, an inverse correlation of Mean Platelet Volume (MPV) with serum CRP level was seen (Nasri et al., 2006b).

In this study, CRP levels above 6.2 had a direct statistically significant correlation with duration of hemodialysis and phosphorus level. CRP levels greater than 10 mg L^-1 have significant relation with age and phosphorus levels (p<0.05). But no significant correlation between CRP and BMI, cholesterol, albumin, hemoglobin, hematocrit and calcium levels was found (p>0.05). On the other hand, studies have reported increased risk of cardiac disease with high CRP levels without a decrease in albumin levels. Therefore, in the presence of normal albumin levels, CRP levels may increase (Razeghi et al., 2008). Absence of a significant correlation between CRP and albumin in the present study was compatible with the study of Nascimento et al. (2004) which observed no significant correlation between CRP and serum albumin level in hemodialysis patients. Also, in another study that conducted on hemodialysis patients, no significant association between serum CRP and serum albumin was seen (Baradaran and Hamid, 2005). Hung et al. (2008) observed a direct relationship between serum phosphorus and CRP in a cohort study In another study, Calo et al. (2011) showed that in dialysis patients, CRP reduction can lead to decrease in phosphorus level.

CONCLUSION

The results of this study demonstrate that the prevalence of inflammation among the hemodialysis patients is high and CRP level greater than 6.2 and 10 mg L^-1 (as a cutoff point), have correlation with age, duration of hemodialysis and phosphorus level in these patients. Considering that even a single determination of CRP level is predictive of poor prognosis in hemodialysis patients, regular checking of CRP may be helpful to tailor appropriate management strategy.

ACKNOWLEDGMENTS

The authors thank of all the participants and the staff of Imam Khomeini and Fatemeh Zahra dialyze centers for their efficient and kind collaboration.