Free serum concentrations of the protein-bound retention solute p-cresol predict mortality in hemodialysis patients
2006; Elsevier BV; Volume: 69; Issue: 6 Linguagem: Inglês
10.1038/sj.ki.5000115
ISSN1523-1755
AutoresBert Bammens, Pieter Evenepoel, H. Keuleers, Kristin Verbeke, Yves Vanrenterghem,
Tópico(s)Muscle metabolism and nutrition
ResumoBased on in vitro data, protein-bound uremic retention solutes have increasingly been recognized to play a pathophysiological role in the uremic syndrome. p-Cresol, a representative of this group of molecules, has been shown to be implicated in uremic immunodeficiency and endothelial dysfunction, potentially linking its serum levels to mortality. Thus far, however, no clinical information on this issue is available. To determine the relationship between p-cresol and all-cause mortality, 175 prevalent hemodialysis (HD) patients were enrolled in a prospective study. At baseline, serum levels of the water-soluble solutes urea, creatinine, and phosphate, the middle molecule β2-microglobulin, total and free concentrations of the protein-bound solute p-cresol, and several risk factors for mortality were evaluated. During a median follow-up of 34 months, 60 patients died. Baseline comorbidity (Davies score) (hazard ratio (HR), 1.49; 95% confidence interval (95% CI), 1.19–1.86), impaired nutritional status (HR, 4.22; 95% CI, 2.15–8.29), time since initiation of dialysis (HR, 0.98; 95% CI, 0.97–1.00), and higher free concentrations of the protein-bound solute p-cresol (HR, 2.28; 95% CI, 1.12–4.64) were independently associated with mortality (multivariate Cox proportional hazards analysis). Our data suggest that free serum levels of p-cresol, a representative of the protein-bound uremic retention solutes, are associated with mortality in HD patients. These findings may encourage nephrologists to widen their field of interest beyond the scope of small water-soluble uremic solutes and middle molecules. Based on in vitro data, protein-bound uremic retention solutes have increasingly been recognized to play a pathophysiological role in the uremic syndrome. p-Cresol, a representative of this group of molecules, has been shown to be implicated in uremic immunodeficiency and endothelial dysfunction, potentially linking its serum levels to mortality. Thus far, however, no clinical information on this issue is available. To determine the relationship between p-cresol and all-cause mortality, 175 prevalent hemodialysis (HD) patients were enrolled in a prospective study. At baseline, serum levels of the water-soluble solutes urea, creatinine, and phosphate, the middle molecule β2-microglobulin, total and free concentrations of the protein-bound solute p-cresol, and several risk factors for mortality were evaluated. During a median follow-up of 34 months, 60 patients died. Baseline comorbidity (Davies score) (hazard ratio (HR), 1.49; 95% confidence interval (95% CI), 1.19–1.86), impaired nutritional status (HR, 4.22; 95% CI, 2.15–8.29), time since initiation of dialysis (HR, 0.98; 95% CI, 0.97–1.00), and higher free concentrations of the protein-bound solute p-cresol (HR, 2.28; 95% CI, 1.12–4.64) were independently associated with mortality (multivariate Cox proportional hazards analysis). Our data suggest that free serum levels of p-cresol, a representative of the protein-bound uremic retention solutes, are associated with mortality in HD patients. These findings may encourage nephrologists to widen their field of interest beyond the scope of small water-soluble uremic solutes and middle molecules. When in the 1960s dialysis strategies as a maintenance therapy for stage 5 chronic kidney disease were introduced in clinical practice, the quality of life and survival of patients improved substantially. Efficiency of dialysis was estimated based on the kinetics of small water-soluble uremic retention solutes, notably urea nitrogen.1.Sargent J.A. Control of dialysis by a single-pool urea model: the National Cooperative Dialysis Study.Kidney Int. 1983; 23: S19-S25Google Scholar The findings of early studies were supportive for this approach, since they indicated a benefit of higher removal of urea nitrogen with regard to morbidity and mortality.2.Parker III, T.F. Husni L. Huang W. et al.Survival of hemodialysis patients in the United States is improved with a greater quantity of dialysis.Am J Kidney Dis. 1994; 23: 670-680Abstract Full Text PDF PubMed Scopus (243) Google Scholar However, despite less efficient removal of small water-soluble molecules by peritoneal dialysis as compared to hemodialysis (HD), peritoneal dialysis patients were seen to have a similar clinical condition as their HD counterparts.3.Tenckhoff H. Curtis F.K. Experience with maintenance peritoneal dialysis in the home.Trans Am Soc Artif Intern Organs. 1970; 16: 90-95PubMed Google Scholar This observation questioned the importance of small water-soluble molecules in the uremic syndrome. It gave rise to the middle-molecule hypothesis,4.Babb A.L. Ahmad S. Bergström J. Scribner B.H. The middle molecule hypothesis in perspective.Am J Kidney Dis. 1981; 1: 46-50Abstract Full Text PDF PubMed Scopus (82) Google Scholar which suggests a pathophysiological role for molecules with a higher molecular mass than urea nitrogen in the development of the uremic syndrome. Nevertheless, since it was impossible at that time to identify these solutes, urea kinetic modeling became established as the most valuable tool for assessment of dialysis adequacy.5.National Kidney Foundation K/DOQI clinical practice guidelines for hemodialysis adequacy, 2000.Am J Kidney Dis. 2001; 37: S7-S64Google Scholar During the previous decades, the middle-molecule hypothesis has regained interest by the characterization of several representatives of this molecular mass class.6.Vanholder R. De Smet R. Glorieux G. et al.Review on uremic toxins: classification, concentration, and interindividual variability.Kidney Int. 2003; 63: 1934-1943Abstract Full Text Full Text PDF PubMed Scopus (1128) Google Scholar Moreover, recent research work has directed attention towards another distinct group of uremic retention molecules: the protein-bound solutes.7.Vanholder R. De Smet R. Lameire N. Protein-bound uremic solutes: the forgotten toxins.Kidney Int. 2001; 59: S266-S270Abstract Full Text PDF Scopus (103) Google Scholar Although most of them have a low molecular mass, binding to serum proteins hampers their dialytic removal. p-Cresol, a 108-Da colonic fermentation metabolite of the amino acid tyrosine, is considered a prototype of this group of uremic solutes. The protein binding of this molecule was shown to be about 90% in stage 5 chronic kidney disease patients.8.De Smet R. David F. Sandra P. et al.A sensitive HPLC method for the quantification of free and total p-cresol in patients with chronic renal failure.Clin Chim Acta. 1998; 278: 1-21Crossref PubMed Scopus (75) Google Scholar Among different in vitro studies, several have suggested that p-cresol plays a role in the immunodeficiency of uremia.9.Vanholder R. De Smet R. Waterloos M.A. et al.Mechanisms of uremic inhibition of phagocyte reactive species production: characterization of the role of p-cresol.Kidney Int. 1995; 47: 510-517Abstract Full Text PDF PubMed Scopus (129) Google Scholar, 10.Wratten M.L. Tetta C. De Smet R. et al.Uremic ultrafiltrate inhibits platelet-activating factor synthesis.Blood Purif. 1999; 17: 134-141Crossref PubMed Scopus (38) Google Scholar, 11.Dou L. Cerini C. Brunet P. et al.p-Cresol, a uremic toxin, decreases endothelial cell response to inflammatory cytokines.Kidney Int. 2002; 62: 1999-2009Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar Recently, a link between p-cresol and endothelial dysfunction has been demonstrated.12.Dou L. Bertrand E. Cerini C. et al.The uremic solutes p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair.Kidney Int. 2004; 65: 442-451Abstract Full Text Full Text PDF PubMed Scopus (344) Google Scholar, 13.Cerini C. Dou L. Anfosso F. et al.p-Cresol, a uremic retention solute, alters the endothelial barrier function in vitro.Thromb Haemost. 2004; 92: 140-150PubMed Google Scholar In spite of these in vitro findings, only few data on the clinical importance of the solute have been reported to date. Free serum concentrations of p-cresol were shown to be higher in HD patients hospitalized for infectious disease.14.De Smet R. Van Kaer J. Van Vlem B. et al.Toxicity of free p-cresol: a prospective and cross-sectional analysis.Clin Chem. 2003; 49: 470-478Crossref PubMed Scopus (103) Google Scholar Furthermore, a positive relationship was found between total serum levels of p-cresol and a uremic symptom score in patients treated with peritoneal dialysis, whereas a correlation with small water-soluble solutes and the middle molecule β2-microglobulin was absent.15.Bammens B. Evenepoel P. Verbeke K. Vanrenterghem Y. Removal of middle molecules and protein-bound solutes by peritoneal dialysis and relation with uremic symptoms.Kidney Int. 2003; 64: 2238-2243Abstract Full Text Full Text PDF PubMed Scopus (156) Google Scholar Based on the above-mentioned in vitro relationships of p-cresol with immune and endothelial cell dysfunction, it is tempting to speculate on this solute as a contributor to the high mortality of dialysis patients. Thus far, however, no information on this issue is available. The principal aim of this prospective observational cohort study was to investigate the association between total and free serum levels of p-cresol and survival in prevalent HD patients. Baseline characteristics of the patients are shown in the first column of Table 1. After a follow-up of 30.1±0.4 months (median 34.0, minimum–maximum 22.0–34.0), 60 of the 175 patients (34%) had died. Causes of death were cardiovascular in 30, infectious in 12, and malignancy in four. Of the remaining 14, one patient died due to a hemorrhagic complication after abdominal surgery and one of noninfectious respiratory insufficiency due to end-stage pulmonary emphysema. In six patients cause of death was defined as cachexia without evidence for malignancy or infection. Six patients died at home from a nonspecified cause. In all, 28 patients were transplanted and six were transferred to another dialysis centre during follow-up.Table 1Baseline characteristics of the patientsVariableAll (n=175)Free p-cresol ≥1.97 mg/l (n=88)Free p-cresol <1.97 mg/l (n=87)PAge (years)64.7±1.1 (67, 26–89)68.2±1.6 (72, 26–89)61.2±1.6 (62, 31–86)0.001Sex (male/female)108/6750/3858/290.180Time since dialysis initiation (months)29.8±2.5 (18.1, 2.3–158.1)37.8±4.2 (24.6, 2.3–158.1)21.7±2.2 (14.9, 2.5–128.0)0.021Treatment modality (HD/HDF)107/6863/2544/430.004Bicarbonate (mEq/l)22.9±0.2 (22.6, 17.2–40.0)22.7±0.2 (22.5, 17.7–26.7)23.1±0.3 (22.8, 17.2–40.0)0.642Hemoglobin (g/dl)11.6±0.1 (11.6, 7.3–15.0)11.5±0.1 (11.5, 7.3–14.8)11.6±0.2 (11.8, 8.0–15.0)0.285Albumin (g/dl)3.64±0.03 (3.66, 2.69–4.48)3.58±0.03 (3.61, 4.37–4.48)3.70±0.04 (3.72, 2.85–4.48)0.017CRP (mg/l)21.8±2.6 (8.0, 3.0–211.8)20.8±3.4 (8.3, 3.0–211.8)22.8±3.9 (7.5, 3.0–184.0)0.564Urea (mg/dl)142.4±3.0 (138, 34–266)155.9±4.4 (152, 63–266)128.8±3.6 (129, 34–244)<0.0001Creatinine (mg/dl)8.6±0.2 (8.6, 2.5–14.5)8.9±0.3 (8.9, 2.8–14.0)8.3±0.3 (8.2, 2.5–14.5)0.097Phosphate (mg/dl)4.8±0.1 (4.6, 1.6–9.3)4.7±0.2 (4.6, 1.9–9.3)5.0±0.2 (4.8, 1.6–9.3)0.309β2-Microglobulin (mg/l)27.7±0.8 (26.5, 8.2–84.9)30.4±0.9 (29.5, 15.0–63.1)24.9±1.1 (23.6, 8.2–84.9)<0.0001Total p-cresol (mg/l)19.0±0.9 (18.9, 0.3–60.5)26.5±1.0 (23.4, 12.3–60.5)11.4±0.8 (13.4, 0.3–25.0)<0.0001Free p-cresol (mg/l)2.59±0.17 (1.97, 0.30–12.74)4.12±0.24 (3.39, 1.97–12.74)1.06±0.05 (1.08, 0.30–1.94)<0.0001% free p-cresol of total13.03±0.44 (11.61, 4.61–39.59)15.49±0.58 (14.25, 7.61–39.59)9.70±0.42 (8.89, 4.61–23.58)<0.0001spKt/V1.64±0.04 (1.50, 0.04–3.80)1.66±0.06 (1.56, 0.04–3.62)1.61±0.05 (1.49, 0.76–3.80)0.276rKt/V0.06±0.01 (0, 0–0.79)0.04±0.01 (0, 0–0.79)0.07±0.01 (0.02, 0–0.43)0.019nPCR (g/kg/day)0.93±0.02 (0.92, 0.15–1.71)1.02±0.03 (1.03, 0.15–1.71)0.85±0.02 (0.83, 0.45–1.51)<0.0001Body weight (kg)66.6±1.1 (64.6, 42.5–112.0)65.9±1.6 (63.2, 45.4–112.0)67.2±1.5 (66.1, 42.5–106.5)0.336BMI (kg/m2)23.9±0.4 (23.0, 16.0–48.6)24.2±0.6 (23.1, 16.9–48.6)23.7±0.4 (22.7, 16.0–35.2)0.929Fat mass (kg)18.4±0.9 (15.3, 3.4–56.3)19.1±1.4 (15.4, 3.4–56.3)17.4±1.2 (15.1, 5.8–40.0)0.631FFM (kg)48.6±0.9 (49.3, 30.3–71.1)47.9±1.2 (49.0, 32.9–64.9)49.5±1.4 (51.1, 30.3–71.1)0.421MAMC (cm)23.4±0.3 (23.2, 15.0–31.4)23.3±0.4 (22.9, 17.0–31.4)23.6±0.5 (23.8, 15.0–30.8)0.362SGA (% impaired nutritional status)32.9%44.0%21.1%0.003MIS6.17±0.33 (5, 0–22)6.95±0.47 (6, 0–22)5.37±0.45 (4, 0–22)0.011Davies score1.84±0.11 (2, 0–6)1.95±0.14 (2, 0–6)1.72±0.16 (2, 0–5)0.246Davies grade (low/medium/high)38/81/5613/47/2825/34/280.053Atherosclerotic disease (%)46.6%54.0%39.0%0.048Diabetes (%)29.7%33.0%26.4%0.346HD, hemodialysis; HDF, hemodiafiltration; CRP, C-reactive protein; spKt/V, single-pool Kt/V of urea nitrogen; rKt/V, renal Kt/V of urea nitrogen; nPCR, normalized protein catabolic rate; BMI, body mass index; FFM, fat-free mass; MAMC, mid-arm muscle circumference; SGA, Subjective Global Assessment; MIS, Malnutrition–Inflammation Score.Data are expressed as mean±s.e.m. (median, minimum–maximum). Differences between patients with high (≥1.97 mg/l) and low (<1.97 mg/l) free p-cresol were evaluated using Mann–Whitney U-test or the χ2 test for association where appropriate. Two-sided P-values are indicated in the last column. Open table in a new tab HD, hemodialysis; HDF, hemodiafiltration; CRP, C-reactive protein; spKt/V, single-pool Kt/V of urea nitrogen; rKt/V, renal Kt/V of urea nitrogen; nPCR, normalized protein catabolic rate; BMI, body mass index; FFM, fat-free mass; MAMC, mid-arm muscle circumference; SGA, Subjective Global Assessment; MIS, Malnutrition–Inflammation Score. Data are expressed as mean±s.e.m. (median, minimum–maximum). Differences between patients with high (≥1.97 mg/l) and low (<1.97 mg/l) free p-cresol were evaluated using Mann–Whitney U-test or the χ2 test for association where appropriate. Two-sided P-values are indicated in the last column. Table 2 shows the relative risks of death estimated by univariate Cox proportional hazards analysis. Age at baseline, serum albumin, C-reactive protein (CRP), creatinine, phosphate, residual renal function, impaired nutritional status according to Subjective Global Assessment (SGA), Malnutrition–Inflammation Score (MIS), Davies score and grade, atherosclerotic disease, and diabetes were significantly associated with mortality. Total serum levels of p-cresol were not associated with outcome. Free concentrations of the solute, however, were significantly related to mortality. Figure 1 shows the Kaplan–Meier survival curves of patients with high (≥1.97 mg/l, n=88) and with low (<1.97 mg/l, n=87) free serum levels of p-cresol. The survival difference between the two groups was statistically significant (log-rank P-value 0.041).Table 2Univariate Cox proportional hazards analysis for association of baseline variables with all-cause mortality (n=175)VariableUnit of increaseHRPAge1 year1.04 (1.02–1.07)0.0002SexMale vs female0.65 (0.39–1.08)0.096Time since dialysis initiation1 month1.00 (0.99–1.01)0.717Treatment modalityHDF vs HD0.58 (0.33–1.04)0.065Bicarbonate1 mEq/l0.98 (0.89–1.08)0.685Hemoglobin1 g/dl0.86 (0.72–1.04)0.123Albumin0.1 g/dl0.89 (0.82–0.96)0.004<3.5 g/dl vs ≥3.5 g/dl1.76 (1.06–2.94)0.030CRP10 mg/l1.07 (1.01–1.13)0.024≥8.0 mg/l vs <8.0 mg/l2.00 (1.17–3.40)0.011Urea1 mg/dl1.00 (0.99–1.01)0.667Creatinine1 mg/dl0.79 (0.70–0.88)<0.0001Phosphate1 mg/dl0.83 (0.70–0.99)0.036β2-microglobulin10 mg/l1.05 (0.83–1.33)0.678Total p-cresol10 mg/l1.17 (0.95–1.46)0.142Free p-cresol1 mg/l1.10 (1.00–1.21)0.058≥1.97 vs <1.97 mg/l1.73 (1.02–2.94)0.043% Free p-cresol of total1%1.03 (0.98–1.07)0.270spKt/V1 unit0.71 (0.38–1.33)0.287rKt/V1 unit0.01 (0.00–1.19)0.060Present vs absent0.51 (0.30–0.88)0.016nPCR1 g/kg/day0.84 (0.28–2.58)0.762Body weight1 kg1.00 (0.98–1.02)0.852BMI1 kg/m21.02 (0.96–1.08)0.465Fat mass1 kg1.02 (0.99–1.05)0.287FFM1 kg0.98 (0.94–1.01)0.127MAMC1 cm0.90 (0.81–1.01)0.068SGA‘Impaired’ vs ‘normal’3.29 (1.87–5.80)<0.0001MIS1 unit1.11 (1.05–1.19)0.001Davies score1 unit1.63 (1.36–1.96)<0.0001Davies grade1 grade2.65 (1.74–4.04)<0.0001Atherosclerotic diseasePresent vs absent2.19 (1.29–3.74)0.004DiabetesPresent vs absent2.37 (1.43–3.94)0.001HD, hemodialysis; HDF, hemodiafiltration; CRP, C-reactive protein; spKt/V, single pool Kt/V of urea nitrogen; rKt/V, renal Kt/V of urea nitrogen; nPCR, normalized protein catabolic rate; BMI, body mass index; FFM, fat-free mass; MAMC, mid-arm muscle circumference; SGA, Subjective Global Assessment; MIS, Malnutrition–Inflammation Score.Relative risks of death for each variable are given as HR (95% CI) and P-value for the univariate Cox proportional hazards analysis. Open table in a new tab HD, hemodialysis; HDF, hemodiafiltration; CRP, C-reactive protein; spKt/V, single pool Kt/V of urea nitrogen; rKt/V, renal Kt/V of urea nitrogen; nPCR, normalized protein catabolic rate; BMI, body mass index; FFM, fat-free mass; MAMC, mid-arm muscle circumference; SGA, Subjective Global Assessment; MIS, Malnutrition–Inflammation Score. Relative risks of death for each variable are given as HR (95% CI) and P-value for the univariate Cox proportional hazards analysis. The second and third columns of Table 1 show baseline data of patients with high and low free concentrations of p-cresol separately. As compared to the latter, those with high levels had lower serum albumin, higher total p-cresol concentrations, and lower renal Kt/V of urea nitrogen (rKt/V). Moreover, they were older, had been treated with dialysis for a longer time, and were more likely to be treated with HD instead of hemodiafiltration (HDF). While their daily protein intake per kg body weight (normalized protein catabolic rate) was higher, they showed more signs of impaired nutritional status based on SGA and had a higher MIS. Finally, a significantly larger proportion of patients with high free levels of p-cresol at baseline had atherosclerotic disease as compared to those with low levels. To identify independent determinants of free p-cresol concentrations, a multivariate linear regression analysis was performed as outlined in Materials and Methods. Since MIS is mathematically dependent on other relevant variables (albumin and SGA), it was not introduced into the model. Also, Davies score, Davies grade, diabetes, and atherosclerotic disease are by definition covarying variables. For that reason, only Davies score was used. In the final regression model, the following variables were retained as independent predictors of the free serum level of p-cresol: serum albumin (P<0.0001), age (P=0.013), total p-cresol (P<0.0001), and impaired nutritional status based on SGA (P=0.005). The respective regression coefficients allowed formulating the following equation, which explained 77.6% (adjusted R2) of the variability of free p-cresol (mg/l): 4.92–1.30 × (serum albumin (g/dl))–0.02 × (age (years))+0.17 × (total p-cresol (mg/l))+0.60 (in case of impaired nutritional status according to SGA). Based on the findings of the univariate Cox proportional hazards analyses (Table 2) and taking the variables related to the free concentration of p-cresol into account (Table 1), the following parameters were entered in the multivariate survival analysis: age, sex, time since initiation of dialysis, treatment modality, hemoglobin, albumin, CRP, urea, creatinine, phosphate, β2-microglobulin, free p-cresol, rKt/V, normalized protein catabolic rate, fat-free mass, mid-arm muscle circumference, SGA, and Davies score. Free p-cresol was entered as a binary variable (≥1.97 mg/l or <1.97 mg/l) (Model A) or as a continuous variable (Model B). The proportionality assumption was fulfilled for all of the variables studied (proportionality test overall P-value=0.536). In Table 3 the final results of Models A and B are given. Time since initiation of dialysis, Davies score, and impaired nutritional status based on SGA were consistently retained as independent factors influencing survival. Depending on the definition of the variable, the free serum level of p-cresol remained in the final model (Model A) or lost significance in the final step of the initial screening (Model B).Table 3Multivariate Cox proportional hazards analysis of baseline variables associated with all-cause mortality (n=175)aSee Materials and methods for detailed description of the statistical methods used.VariableUnit of increaseHR (95% CI)PModel A (‘free p-cresol’ entered as a binary variable) Months on dialysis1 month0.98 (0.97–1.00]0.005 Free p-cresolhigh vs lowb‘High’ means ≥1.97 mg/l, ‘low’ means <1.97 mg/l.2.28 (1.12–4.64)0.023 SGA‘impaired’ vs ‘normal’4.22 (2.15–8.29)<0.0001 Davies score1 unit1.49 (1.19–1.86)0.001Model B (‘free p-cresol’ entered as a continuous variable) Age1 year1.03 (1.00–1.06)0.047 Months on dialysis1 month0.98 (0.97–1.00)0.007 SGA‘impaired’ vs ‘normal’4.42 (2.25–8.71)<0.0001 Davies score1 unit1.45 (1.17–1.80)0.001SGA, Subjective Global Assessment.Relative risk of death for each variable is given as HR (95% CI) and P-value for Cox proportional hazards analysis.a See Materials and methods for detailed description of the statistical methods used.b ‘High’ means ≥1.97 mg/l, ‘low’ means <1.97 mg/l. Open table in a new tab SGA, Subjective Global Assessment. Relative risk of death for each variable is given as HR (95% CI) and P-value for Cox proportional hazards analysis. The results of this prospective observational cohort study suggest that, besides other well-known predictors of survival, the free serum level of the protein-bound uremic retention solute p-cresol is related to mortality in patients treated with dialysis. Several in vitro data have suggested a pathophysiological role of the solute in some important aspects of the uremic syndrome. p-Cresol depressed the respiratory burst reactivity of granulocytes and monocytes at concentrations encountered in stage 5 chronic kidney disease patients.9.Vanholder R. De Smet R. Waterloos M.A. et al.Mechanisms of uremic inhibition of phagocyte reactive species production: characterization of the role of p-cresol.Kidney Int. 1995; 47: 510-517Abstract Full Text PDF PubMed Scopus (129) Google Scholar Later it was shown that the solute inhibits the synthesis of platelet-activating factor, another aspect of leukocyte function.10.Wratten M.L. Tetta C. De Smet R. et al.Uremic ultrafiltrate inhibits platelet-activating factor synthesis.Blood Purif. 1999; 17: 134-141Crossref PubMed Scopus (38) Google Scholar Further evidence for a role of p-cresol in the immunodeficiency of uremia was provided by the recent finding of its inhibitory effect on cytokine-induced expression of the endothelial adhesion molecules intercellular adhesion molecule-1 and vascular cellular adhesion molecule-1.11.Dou L. Cerini C. Brunet P. et al.p-Cresol, a uremic toxin, decreases endothelial cell response to inflammatory cytokines.Kidney Int. 2002; 62: 1999-2009Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar Endothelial dysfunction is another characteristic of the uremic syndrome.16.Annuk M. Mihkel Z. Lind L. et al.Oxidative stress and endothelial function in chronic renal failure.J Am Soc Nephrol. 2001; 12: 2747-2752PubMed Google Scholar It is considered to be an important contributor to the burden of cardiovascular atherosclerotic complications in renal disease patients. That p-cresol might be implicated in this disorder was suggested recently by two in vitro studies that described an inhibition of the regenerative properties and an increase in the permeability of endothelial cells by uremic concentrations of the solute.12.Dou L. Bertrand E. Cerini C. et al.The uremic solutes p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair.Kidney Int. 2004; 65: 442-451Abstract Full Text Full Text PDF PubMed Scopus (344) Google Scholar, 13.Cerini C. Dou L. Anfosso F. et al.p-Cresol, a uremic retention solute, alters the endothelial barrier function in vitro.Thromb Haemost. 2004; 92: 140-150PubMed Google Scholar In accordance with the data linking p-cresol to immunodeficiency, a study by De Smet et al.14.De Smet R. Van Kaer J. Van Vlem B. et al.Toxicity of free p-cresol: a prospective and cross-sectional analysis.Clin Chem. 2003; 49: 470-478Crossref PubMed Scopus (103) Google Scholar described higher free serum levels of the solute in a group of HD patients hospitalized for infectious disease as compared to those hospitalized for other reasons. From the absence of a difference in total p-cresol between the two groups and results of in vitro experiments described in the same paper, the authors concluded that, similar to protein-bound drugs, the free, rather than the total, concentration of p-cresol is responsible for its biological activities. Our findings are in agreement with that thesis. Indeed, although survival of patients with total serum levels of p-cresol ≥18.9 mg/l was worse than in those with lower levels (50 vs 66%, log-rank P-value=0.456), only the free serum concentrations of the solute conferred significantly to outcome. Instead of attributing our findings to the sole influence of p-cresol, it seems more reasonable to consider the molecule as a representative of the group of protein-bound retention solutes. Indeed, also other members of this group have been found to be associated with immune dysfunction, endothelial cell dysfunction, and, closely related to the latter, oxidative stress. Like p-cresol, indoxyl sulfate was found to inhibit endothelial proliferation and wound repair.12.Dou L. Bertrand E. Cerini C. et al.The uremic solutes p-cresol and indoxyl sulfate inhibit endothelial proliferation and wound repair.Kidney Int. 2004; 65: 442-451Abstract Full Text Full Text PDF PubMed Scopus (344) Google Scholar Moreover, the molecule was shown to induce free radical production by renal tubular cells and activate nuclear factor κB, a central mediator of atherosclerosis.17.Motojima M. Hosokawa A. Yamato H. et al.Uremic toxins of organic anions up-regulate PAI-1 expression by induction of NF-κB and free radical in proximal tubular cells.Kidney Int. 2003; 63: 1671-1680Abstract Full Text Full Text PDF PubMed Scopus (196) Google Scholar Recently, another protein-bound uremic compound, phenylacetic acid, was identified to have an inhibitory impact on inducible nitric oxide synthase, an enzyme with a protective potential on endothelial function.18.Jankowski J. van der Giet M. Jankowski V. et al.Increased plasma phenylacetic acid in patients with end-stage renal failure inhibits iNOS expression.J Clin Invest. 2003; 112: 256-264Crossref PubMed Scopus (94) Google Scholar In accordance with earlier observations, our data confirm that the relative risk of death in dialysis patients increases with higher age, the presence of atherosclerotic disease, diabetes, and higher overall comorbidity (Table 2).19.Eknoyan G. Beck G. Cheung A. et al.Effect of dialysis dose and membrane flux in maintenance hemodialysis.N Engl J Med. 2002; 347: 2010-2019Crossref PubMed Scopus (1085) Google Scholar Residual renal function seems to be a protective factor. Although the latter has been shown before in several studies including peritoneal dialysis patients, only few other studies with HD patients report on this issue.20.Bargman J.M. Thorpe K.E. Churchill D.N. for the CANUSA PD Study Group Relative contribution of residual renal function and peritoneal clearance to adequacy of dialysis: a reanalysis of the CANUSA study.J Am Soc Nephrol. 2001; 12: 2158-2162PubMed Google Scholar, 21.Termorshuizen F. Dekker F.W. van Manen J.G. et al.Relative contribution of residual renal function and different measures of adequacy to survival in hemodialysis patients: an analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD)-2.J Am Soc Nephrol. 2004; 15: 1061-1070Crossref PubMed Scopus (259) Google Scholar Finally, parameters reflecting impaired nutritional status and/or inflammation, such as low serum albumin, low creatinine, high CRP, unfavorable SGA, and high MIS, are strong predictors of mortality. These observations are in agreement with literature data reporting on the association between malnutrition, inflammation, atherosclerosis, and mortality in renal disease patients.22.Pecoits-Filho R. Lindholm B. Stenvinkel P. The malnutrition, inflammation, and atherosclerosis (MIA) syndrome – the heart of the matter.Nephrol Dial Transplant. 2002; 17: 28-31Crossref PubMed Scopus (326) Google Scholar In each of the multivariate models, comorbidity and nutritional status were retained as independent factors influencing survival (Table 3). Depending on the definition of the variable, the free serum level of p-cresol remained in the model or lost its significance in one of the final elimination steps. This discrepancy between Models A and B may seem confusing at first sight. In our opinion, however, it reflects the complex interplay between (free) p-cresol and other important baseline variables as far as their relationship with outcome is concerned. As can be appreciated from Table 1 and from the multivariate regression analysis, baseline free serum concentrations of p-cresol were indeed significantly related to age, serum albumin, impaired nutritional status according to SGA, MIS, and the presence of atherosclerotic disease, all associated with mortality themselves. The present study, however, does not allow to define the exact role of (free) p-cresol within this group of variables and further investigation will be needed to resolve this intriguing question. Nevertheless, whatever the strength and causality of the relationship between p-cresol and the other variables, it is clear from our data that a high free serum level of the solute (or related solutes) constitutes an additive risk factor for mortality in a dialysis population. Furthermore, it should be noted that only higher free levels of
Referência(s)