Modulation and Source of Procalcitonin in ... - Wiley Online Library

..................................................................................................................................................................................................

Modulation and Source of Procalcitonin in Reduced Renal Function and Renal Replacement Therapy S. Herget-Rosenthal*, T. Klein*, G. Marggrafy, T. Hirschz, H.-G. Jakoby, T. Philipp* & A. Kribben*

Abstract *Department of Nephrology; yDepartment of Thoracic and Cardiovascular Surgery; and zDepartment of Trauma Surgery, Section Surgical Research, University Hospital, Essen, Germany Received 7 October 2004; Accepted in revised form 5 November 2004 Correspondence to: Dr S. Herget-Rosenthal, MD, Klinik fu¨r Nieren- und Hochdruckkrankheiten, Universita¨tsklinikum Essen, Hufelandstr. 55, D-45122 Essen, Germany. E-mail: [email protected]

Serum procalcitonin (PCT), an accurate marker of severe infection, is moderately increased in chronic kidney disease (CKD), peritoneal dialysis (PD) and haemodialysis (HD). We studied the extent of PCT elevation and factors accounting for elevated PCT in CKD and dialysis, and whether peripheral blood mononuclear cells (PBMC) contribute to increased PCT. In 37 controls, 281 CKD, 31 PD, and 65 HD patients without infection, PCT was measured and correlated with CKD stage, PD, HD, C-reactive protein (CRP), cardiovascular disease (CVD) and other clinical parameters. PCT release by PBMC from controls, advanced CKD, PD and HD patients (12 subjects each) was measured. PCT increased in parallel to the deterioration of CKD. Oliguria, advanced CKD, PD, HD, CVD and elevated CRP were independently associated with PCT elevation. PCT release from PBMC significantly increased in advanced CKD, PD and HD. PCT release from PBMC correlated closely with the corresponding serum PCT values (r ¼ 0.76, P < 0.001). In the absence of infection, PCT may increase due to reduced renal elimination and increased synthesis, as due to PBMC. Furthermore, serum PCT could serve as a marker of low-grade inflammation and CVD, which substantially increase mortality in CKD and dialysis.

Introduction Procalcitonin (PCT), a 13-kDa protein, is a precursor of calcitonin which is physiologically synthesized by the C cells of the thyroid gland and pulmonary neuroendocrine cells in minute quantities [1, 2]. Serum PCT levels below 0.1 ng/ml are commonly detected in healthy controls [3, 4]. Sepsis, severe bacterial and fungal infections induce a marked increase of serum PCT. Serum PCT has been established as a very accurate and specific marker of severe systemic infection in patients with normal renal function [5–7]. In inflammation and sepsis, PCT synthesis was demonstrated in various tissues and cell types such as leucocytes, especially peripheral blood mononuclear cells (PBMC) [1, 2, 8, 9]. Current findings suggest that PCT may not only be a valid marker of infection and inflammation but also a proinflammatory, cytokine-like mediator [10–13]. Elevation of PCT may be harmful as it augmented mortality in sepsis, and immunoneutralization of PCT improved survival in animal models of sepsis [11]. Recent data demonstrated that serum PCT is also an accurate indicator of severe infection and sepsis in patients with chronic kidney disease (CKD) requiring intermittent haemodialysis (HD) [14, 15]. Moderately increased PCT was observed in CKD with and without renal replacement

180

therapy (RRT) in the absence of infections [14–19]. This is in accordance with reports of PCT elevation in various noninfectious, inflammatory conditions such as active autoimmune disease [17], acute pancreatitis [20], trauma [4], major surgery [21] and systemic inflammatory response syndrome [6, 7]. However, the extent of PCT elevation in CKD with and without RRT, its underlying factors and the origin of PCT under these conditions remain to be elucidated. Given recent reports that PBMC are a potential source of PCT, we hypothesized that in CKD and RRT, PCT may also originate from PBMC [8, 9, 22, 23]. Our aim was twofold – to assess the extent of serum PCT elevation and to identify factors which may account for elevated PCT in CKD and RRT in the absence of infection (Study 1) and to investigate whether PBMC contribute to PCT synthesis in CKD and RRT (Study 2).

Subjects and methods Study 1. Patients: Serum PCT was determined in 37 controls without renal dysfunction, defined as nonpathological proteinuria, normal glomerular filtration rate (GFR) and

# 2005 Blackwell Publishing Ltd. Scandinavian Journal of Immunology 61, 180–186

S. Herget-Rosenthal et al. Procalcitonin Changes in CKD and RRT 181 ..................................................................................................................................................................................................

urine sediment, 281 patients with CKD stages I–V, 31 CKD patients treated with PD and 65 CKD patients treated with HD. CKD was classified according to the recent stages proposed by the Kidney Disease Outcomes Quality Initiative of the National Kidney Foundation [24]: stage 1 – GFR 90, stage II – GFR 60–89, stage III – GFR 30–59, stage IV – GFR 15–29 and stage V – GFR < 15 ml/ min/1.73 m2. CKD stage V was modified to include only patients not treated by RRT. All patients and controls were Caucasians, in stable clinical condition without clinical, radiological or microbiological signs of infection or any other apparent acute inflammatory disorder. We excluded patients and controls with thyroid or bronchial carcinoma, acute myocardial infarction, acute pancreatitis, active vasculitis or systemic autoimmune disease, surgery or severe trauma in the 7 days prior to the study, because these conditions potentially increase PCT [4, 17, 20, 21]. No one was treated with immunosuppressive or nonsteroidal anti-inflammatory drugs [25]. Cardiovascular disease (CVD) was defined by the presence of coronary heart disease (angina or myocardial infarction), cerebrovascular (transient ischaemic attack or stroke) and/or peripheral arterial disease, diagnosed by history, physical examination and the appropriate diagnostic tests [26]. Dialysis procedure: Peritoneal dialysis (PD) and HD patients were dialyzed for at least 6 months prior to the study. Ten PD (33%) and 53 HD patients (82%) were anuric. Twenty-one PD patients performed continuous ambulatory PD on a 4–5 exchanges per day schedule. Ten patients were dialyzed with nightly continuous cyclic PD for 8–11 h and a total dialysate exchange of 16–20 l, and additionally 1–2 daily long dwell exchanges. All PD solutions contained glucose or icodextrin and were lactate buffered (Fresenius Medical Care, Bad Homburg, Germany and Baxter, Unterschleissheim, Germany). HD was performed for 4.3 0.6 h per session, three times per week, with steam-sterilized polysulfone or polyamide high-flux membranes (Fresenius Medical Care and Gambro, Planegg-Martinsried, Germany). All membranes were first used and the dialysate buffer was bicarbonate. The dialysate was prepared using tap water and purified by reversed osmosis. Microbiological testing of the dialysate showed no bacterial contamination (0 colony-forming units/ml) and endotoxine levels 0.26 ng/ml and >1.5 mg/l, respectively) were considered elevated for uni- and multivariate analysis. Potential risk factors for elevated PCT were coded as present or absent and assessed by univariate analysis. Variables with a P < 0.20 were included in the multivariate, logistic regression analysis. A P value of less than 0.05 was considered to be statistically significant. Patient and dialysis data were obtained from hospital records. The study protocol is in accordance with the Helsinki Declaration of 1975 as revised in 1996. Informed consent was obtained from all patients prior to enrolment.

Results Serum PCT in CKD and RRT

Characteristics of patients and controls in study 1 are presented in Table 1. The groups studied were similar in respect to age and gender. There were significantly more HD patients with diabetes (Table 1). Characteristics of patients and controls in study 2 did not substantially differ from those in study 1 in regard to age, gender and aetiology of CKD. Renal function of patients and controls in study 1 is presented in Table 2. In study 2, controls, CKD stage V, PD and HD patients had GFR of 106 21, 9 3, 4 2 and 0.9 0.5 ml/min/1.73 m2, respectively. Urine output measured 0.6 0.3 l/d in PD and 0.2 0.1 l/d in HD patients. Thus, PD patients displayed higher residual GFR and urine output than HD patients in both studies. PD patients had also been treated by RRT for a shorter duration compared to HD patients (31 19 versus 47 28 months in study 1, and 25 13 versus 39 25 months in study 2). Serum PCT was in the normal range in controls (Fig. 1). Although elevated, PCT levels in CKD patients at stages I–IV did not markedly differ from those in controls. Serum PCT in CKD at stage V and PD were significantly higher than those of Table 1 Characteristics of controls, chronic kidney disease (CKD) patients stage I–V, and CKD patients with haemodialysis (HD) and peritoneal dialysis (PD) from study 1 CKD aetiology Female (%)

Age (years)

37

38

54 11

0

0

0

29 62 105 53 32 31 65

45 48 42 44 51 48 43

51 15 50 13 47 16 53 15 57 20 50 12 59 18

23 21 13 19 13 16 34*

17 15 20 17 16 7 12

60 64 67 64 72 77 54

n Controls CKD I II III IV V PD HD

Diabetes (%)

Vascular (%)

Other (%)

*P < 0.01 versus CKD III and V and PD, P < 0.05 versus CKD IV.

controls and of the CKD stages I–IV (P < 0.05). HD was associated with significantly higher PCT values than any CKD stage and PD. Serum PCT and GFR in controls, CKD, PD and HD correlated inversely (r ¼ 0.65; P < 0.001). In HD and PD, there was a trend towards higher serum PCT levels in patients without compared to those with residual urine output, which was not statistically significant (0.84 0.41 versus 0.76 0.35 ng/ml in HD and 0.50 0.21 versus 0.43 0.18 ng/ml, respectively; both n.s). Although CRP levels in CKD patients stages I–IV were low and did not significantly vary from controls, there was a tendency that CRP mildly increased with progression of CKD (Table 2). In parallel to PCT, a moderate elevation of CRP was observed in CKD stage V. Serum CRP further increased in PD and especially in HD patients. Serum PCT and serum CRP were positively correlated (r ¼ 0.73; P < 0.001). In study 1 and 2, the findings regarding serum PCT and CRP in controls, CKD stage V, PD and HD were consistent. With progression of CKD, the prevalence of CVD increased with a maximum at CKD stage V (Table 2). PD was associated with a further increase, and HD had the highest rate of CVD. When dividing each subgroup into patients with and without CVD, we observed significantly higher serum PCT values in patients with CVD in CKD stage V, PD and HD compared with patients without CVD (Table 2). Factors associated with elevated serum PCT

As shown in Table 3, patients with PCT above 0.26 ng/ml, which was considered to be elevated for univariate analysis, more often presented with diabetes, CKD stage V, oliguria, defined as a urine output of less than 400 ml per day, PD, HD, longer duration of RRT treatment, CVD and CRP above 1.5 mg/l. Multivariate analysis identified CKD stage V and oliguria as parameters of renal function, PD and HD as renal replacement therapies, presence of CVD and elevated serum CRP as independent factors which were significantly associated with elevated serum PCT in our cohort (Table 4). PCT release from PBMC in CKD and RRT

In study 2, PCT in the supernatant of PBMC was 0.09 0.01 ng/ml at baseline in controls. In CKD, PBMC were isolated exclusively from patients at stage V, as only these patients differed significantly in regard to serum PCT values from controls. In CKD stage V, PD and HD, baseline PCT values in the supernatant did not differ from controls (0.09 0.01, 0.10 0.01 and 0.09 0.02 ng/ml, respectively). After 24-h incubation, PCT in the supernatant had not markedly changed in controls (0.10 0.02) but had significantly increased in CKD stage V (0.13 0.03 ng/ ml), PD (0.15 0.03 ng/ml) and HD (0.15 0.04 ng/ml), compared to the respective baseline values (all P < 0.05).


S. Herget-Rosenthal et al. Procalcitonin Changes in CKD and RRT 183 .................................................................................................................................................................................................. Table 2 Renal function, C-reactive protein (CRP), serum procalcitonin, and the prevalence of cardiovascular disease (CVD) in controls, in chronic kidney disease stages I–V (CKD I–V), haemodialysis (HD) and peritoneal dialysis (PD) from study 1. Serum PCT values are presented separately for patients with CVDþ and without CVD– Procalcitonin (ng/ml)

Controls CKD I II III IV V PD HD

GFR (ml/min/1.73 m2)

Urine output (l/d)

CRP (mg/l)

CVD (%)

CVDþ

CVD–

112 12

n.a.

0.7 0.4

19

0.08 0.01

0.10 0.03

106 11 75 10 44 8 22 4 10 5 32 0.7 0.6

n.a. n.a. n.a. n.a. n.a. 0.7 0.5 0.1 0.1

0.9 0.7 1.0 0.8 1.3 1.0 1.5 1.1 2.9 1.7 4.1 1.7z 5.3 2.1z

17 15 19 23 34 * 45§ 48{

0.09 0.04 0.10 0.07 0.13 0.09 0.23 0.14 0.51 0.23y 0.56 0.23y 0.92 0.42y

0.10 0.01 0.09 0.02 0.14 0.07 0.14 0.08 0.29 0.21 0.38 0.18 0.64 0.34

GFR, glomerular filtration rate. *P < 0.05 versus controls; yP > 0.05 versus CKD– patients of the respective group; zP < 0.05 versus control and CKD I–III; §P < 0.01 versus controls and CKD II; {P < 0.01 versus controls and P < 0.05 versus CKD I–III.

Finally after 48 h incubation, PCT in the supernatant increased further to 0.16 0.03 ng/ml in CKD stage V to 0.18 0.07 ng/ml in PD, and to 0.20 0.11 ng/ml in HD (all P < 0.05 versus the respective baseline values), while PCT values remained at 0.11 0.02 ng/ml in controls. Normalized for the number of PBMC, the release of PCT into the supernatant during 48 h of incubation is illustrated in Fig. 2. PCT release of PBMC isolated from controls was the lowest. In patients with CKD stage V and in PD, PCT release was significantly higher compared to controls (both P < 0.05). HD patients demonstrated the highest PCT release by PBMC, which significantly differed from controls (P < 0.01), as well as CKD and PD patients (P < 0.05 each). As demonstrated in Fig. 3, a strong correlation was observed between serum PCT and PCT released by PBMC into the supernatant during 48 h of

incubation in a combined analysis of all four cohorts of study 2 (r ¼ 0.76, P < 0.001).

Discussion To our knowledge, this is the first study which shows that in the absence of infection, serum PCT progressively increases in parallel to the deterioration of renal function, characterized by advancing CKD and oliguria. This indicates that the progression of CKD and the decrease of urine output will potentially reduce the renal elimination of PCT, which as a low molecular weight protein is probably cleared by glomerular filtration. The close correlation of serum PCT and GFR observed by us further emphasizes this assumption. Consistent with our findings are previous data that the amount of PCT eliminated via

1.5

Serum procalcitonin [ng/ml]

1.25

†

1

0.75

*

*

CKD V

PD

0.5

0.25

0 Control

CKD I

CKD II

CKD III

CKD IV

HD

Figure 1 Serum procalcitonin in controls, in patients with various stages of chronic kidney disease (CKD I–V), and in CKD patients treated with peritoneal dialysis (PD) and haemodialysis (HD). Columns represent mean SD. *P < 0.05 versus control and CKD I–IV; yP < 0.01 versus control and CKD I–IV and P < 0.05 versus CKD V and PD.


Gender (female) Age > 65 years Diabetes Urine output 48 months Cardiovascular disease C-reactive protein >1.5 mg/l

Elevated PCT (>0.26 ng/ml)

Normal PCT (¼0.26 ng/ml)

n ¼ 172 (%)

n ¼ 242 (%)

P

48.8 41.9 26.7 29.7

43.8 40.5 10.8 2.5

0.36 0.86