Title: Protein-energy wasting modifies the association of ghrelin with inflammation, leptin, and mortality in hemodialysis patients
Abstract: Ghrelin abnormalities contribute to anorexia, inflammation, and cardiovascular risk in hemodialysis patients, leading to worse outcome. However, ghrelin levels are influenced by the nutritional status of the individual. We hypothesized that the consequences of ghrelin alterations in hemodialysis patients are context sensitive and dependent on the presence of protein-energy wasting (PEW). In this cross-sectional study of 217 prevalent hemodialysis patients followed for 31 months, we measured ghrelin, leptin, PEW (subjective global assessment), and C-reactive protein (an index of inflammation). Compared to patients in the middle and upper tertile of ghrelin levels, those in the lowest tertile were older, had higher leptin levels and body mass index, and presented an increased mortality risk that persisted after adjustment for age, gender, and dialysis vintage. This risk was lost after correction for comorbidities. Patients with PEW and low ghrelin values had abnormally high C-reactive protein and leptin by multivariate analysis of variance, and the highest mortality risk compared to non-PEW with high ghrelin from all-cause and cardiovascular-related mortality (adjusted hazard ratios of 3.34 and 3.54, respectively). Low ghrelin values in protein-energy wasted hemodialysis patients were linked to a markedly increased cardiovascular mortality risk. Thus, since these patients were more anorectic, our results provide a clinical scenario where ghrelin therapies may be particularly useful. Ghrelin abnormalities contribute to anorexia, inflammation, and cardiovascular risk in hemodialysis patients, leading to worse outcome. However, ghrelin levels are influenced by the nutritional status of the individual. We hypothesized that the consequences of ghrelin alterations in hemodialysis patients are context sensitive and dependent on the presence of protein-energy wasting (PEW). In this cross-sectional study of 217 prevalent hemodialysis patients followed for 31 months, we measured ghrelin, leptin, PEW (subjective global assessment), and C-reactive protein (an index of inflammation). Compared to patients in the middle and upper tertile of ghrelin levels, those in the lowest tertile were older, had higher leptin levels and body mass index, and presented an increased mortality risk that persisted after adjustment for age, gender, and dialysis vintage. This risk was lost after correction for comorbidities. Patients with PEW and low ghrelin values had abnormally high C-reactive protein and leptin by multivariate analysis of variance, and the highest mortality risk compared to non-PEW with high ghrelin from all-cause and cardiovascular-related mortality (adjusted hazard ratios of 3.34 and 3.54, respectively). Low ghrelin values in protein-energy wasted hemodialysis patients were linked to a markedly increased cardiovascular mortality risk. Thus, since these patients were more anorectic, our results provide a clinical scenario where ghrelin therapies may be particularly useful. The prevalence of protein-energy wasting (PEW), manifested as a loss of muscle mass and a mismatch between energy expenditure and intake, is high in advanced chronic kidney disease (CKD).1.Fouque D. Kalantar-Zadeh K. Kopple J. et al.A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease.Kidney Int. 2008; 73: 391-398Abstract Full Text Full Text PDF PubMed Scopus (1118) Google Scholar, 2.Carrero J.J. Identification of patients with eating disorders: clinical and biochemical signs of appetite loss in dialysis patients.J Ren Nutr. 2009; 19: 10-15Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 3.Avesani C.M. Carrero J.J. Axelsson J. et al.Inflammation and wasting in chronic kidney disease: partners in crime.Kidney Int Suppl. 2006; 70: S8-S13Abstract Full Text Full Text PDF Scopus (64) Google Scholar Like in other wasted patient groups, anorexia is common and often linked to persistent systemic inflammation, reduced quality of life, and increased mortality.4.Carrero J.J. Qureshi A.R. Axelsson J. et al.Comparison of nutritional and inflammatory markers in dialysis patients with reduced appetite.Am J Clin Nutr. 2007; 85: 695-701PubMed Google Scholar,5.Kalantar-Zadeh K. Block G. McAllister C.J. et al.Appetite and inflammation, nutrition, anemia, and clinical outcome in hemodialysis patients.Am J Clin Nutr. 2004; 80: 299-307PubMed Google Scholar The regulation of anorexia includes a complex hypothalamic process in which different appetite-regulating centers are affected not only by neuropeptides, but also by peripheral signals from fat tissue and the gut.6.Carrero J.J. Aguilera A. Stenvinkel P. et al.Appetite disorders in uremia.J Ren Nutr. 2008; 18: 107-113Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar, 7.Bossola M. Tazza L. Giungi S. et al.Anorexia in hemodialysis patients: an update.Kidney Int. 2006; 70: 417-422Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 8.Aguilera A. Codoceo R. Bajo M.A. et al.Eating behavior disorders in uremia: a question of balance in appetite regulation.Semin Dial. 2004; 17: 44-52Crossref PubMed Scopus (50) Google Scholar Ghrelin is an orexigenic peptide released primarily from endocrine cells in the stomach, which increases appetite and adjusts both short-term and long-term energy balance.9.Nagaya N. Kojima M. Uematsu M. et al.Hemodynamic and hormonal effects of human ghrelin in healthy volunteers.Am J Physiol Regul Integr Comp Physiol. 2001; 280: R1483-R1487PubMed Google Scholar The orexigenic effects of ghrelin are mediated through the type 1a growth hormone secretagogue receptor, leading to increased gene expression of orexigenic neuropeptides and increased growth hormone (GH) release.10.Nakazato M. Murakami N. Date Y. et al.A role for ghrelin in the central regulation of feeding.Nature. 2001; 409: 194-198Crossref PubMed Scopus (2815) Google Scholar In advanced CKD, total ghrelin levels are high11.Rodriguez Ayala E. Pecoits-Filho R. Heimburger O. et al.Associations between plasma ghrelin levels and body composition in end-stage renal disease: a longitudinal study.Nephrol Dial Transplant. 2004; 19: 421-426Crossref PubMed Scopus (107) Google Scholar,12.Yoshimoto A. Mori K. Sugawara A. et al.Plasma ghrelin and desacyl ghrelin concentrations in renal failure.J Am Soc Nephrol. 2002; 13: 2748-2752Crossref PubMed Scopus (232) Google Scholar—a finding that seems counterintuitive given its orexigenic action and that has been interpreted as a defense mechanism against starvation. Yet, and despite this elevation/resistance, subcutaneous ghrelin administration resulted in several-fold increases in plasma ghrelin concentration followed by improvements in short-term energy intake and energy balance in mildly to moderately malnourished dialysis patients.13.Wynne K. Giannitsopoulou K. Small C.J. et al.Subcutaneous ghrelin enhances acute food intake in malnourished patients who receive maintenance peritoneal dialysis: a randomized, placebo-controlled trial.J Am Soc Nephrol. 2005; 16: 2111-2118Crossref PubMed Scopus (170) Google Scholar,14.Ashby D.R. Ford H.E. Wynne K.J. et al.Sustained appetite improvement in malnourished dialysis patients by daily ghrelin treatment.Kidney Int. 2009; 76: 199-206Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar Similarly, a superagonist of GH-releasing hormone caused rapid improvement of nutritional status in CKD stage 4 and 5 patients without apparent GH deficiency.15.Niemczyk S. Sikorska H. Wiecek A. et al.A super-agonist of growth hormone-releasing hormone causes rapid improvement of nutritional status in patients with chronic kidney disease.Kidney Int. 2010; 77: 450-458Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar Furthermore, ghrelin appears to be involved in other pathophysiological pathways such as improvement of cardiac function,16.Nagaya N. Moriya J. Yasumura Y. et al.Effects of ghrelin administration on left ventricular function, exercise capacity, and muscle wasting in patients with chronic heart failure.Circulation. 2004; 110: 3674-3679Crossref PubMed Scopus (417) Google Scholar,17.Nagaya N. Uematsu M. Kojima M. et al.Chronic administration of ghrelin improves left ventricular dysfunction and attenuates development of cardiac cachexia in rats with heart failure.Circulation. 2001; 104: 1430-1435Crossref PubMed Scopus (437) Google Scholar suppression of sympathetic activity,18.Ashitani J. Matsumoto N. Nakazato M. Ghrelin and its therapeutic potential for cachectic patients.Peptides. 2009; 30: 1951-1956Crossref PubMed Scopus (33) Google Scholar inhibition of the inflammatory response,19.Li W.G. Gavrila D. Liu X. et al.Ghrelin inhibits proinflammatory responses and nuclear factor-kappaB activation in human endothelial cells.Circulation. 2004; 109: 2221-2226Crossref PubMed Scopus (441) Google Scholar,20.Dixit V.D. Schaffer E.M. Pyle R.S. et al.Ghrelin inhibits leptin- and activation-induced proinflammatory cytokine expression by human monocytes and T cells.J Clin Invest. 2004; 114: 57-66Crossref PubMed Scopus (703) Google Scholar anabolic effects on lean mass,21.Deboer M.D. Zhu X. Levasseur P.R. et al.Ghrelin treatment of chronic kidney disease: improvements in lean body mass and cytokine profile.Endocrinology. 2008; 149: 827-835Crossref PubMed Scopus (123) Google Scholar,22.DeBoer M.D. Zhu X.X. Levasseur P. et al.Ghrelin treatment causes increased food intake and retention of lean body mass in a rat model of cancer cachexia.Endocrinology. 2007; 148: 3004-3012Crossref PubMed Scopus (155) Google Scholar metabolic syndrome,23.Lee C.C. Lee R.P. Subeq Y.M. et al.Fasting serum total ghrelin level inversely correlates with metabolic syndrome in hemodialysis patients.Arch Med Res. 2008; 39: 785-790Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar and mediation in insulin sensitivity signaling24.Poykko S.M. Kellokoski E. Horkko S. et al.Low plasma ghrelin is associated with insulin resistance, hypertension, and the prevalence of type 2 diabetes.Diabetes. 2003; 52: 2546-2553Crossref PubMed Scopus (439) Google Scholar or atherosclerosis.25.Poykko S.M. Kellokoski E. Ukkola O. et al.Plasma ghrelin concentrations are positively associated with carotid artery atherosclerosis in males.J Intern Med. 2006; 260: 43-52Crossref PubMed Scopus (55) Google Scholar In CKD, all these pathways have also been linked to PEW.1.Fouque D. Kalantar-Zadeh K. Kopple J. et al.A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease.Kidney Int. 2008; 73: 391-398Abstract Full Text Full Text PDF PubMed Scopus (1118) Google Scholar Several studies, both in animals and humans, have suggested that not only is ghrelin dependent on body fat mass,26.Tschop M. Weyer C. Tataranni P.A. et al.Circulating ghrelin levels are decreased in human obesity.Diabetes. 2001; 50: 707-709Crossref PubMed Scopus (1645) Google Scholar but is also influenced by the individual's nutritional status; although the orexigenic effect of peripheral ghrelin administration differed between rats with different baseline food intake,27.Gilg S. Lutz T.A. The orexigenic effect of peripheral ghrelin differs between rats of different age and with different baseline food intake, and it may in part be mediated by the area postrema.Physiol Behav. 2006; 87: 353-359Crossref PubMed Scopus (54) Google Scholar ghrelin values were markedly different among women with anorexia nervosa and constitutionally thin women, who display a similar low body mass index (BMI) but no nutritional disorder.28.Tolle V. Kadem M. Bluet-Pajot M.T. et al.Balance in ghrelin and leptin plasma levels in anorexia nervosa patients and constitutionally thin women.J Clin Endocrinol Metab. 2003; 88: 109-116Crossref PubMed Scopus (268) Google Scholar,29.Germain N. Galusca B. Le Roux C.W. et al.Constitutional thinness and lean anorexia nervosa display opposite concentrations of peptide YY, glucagon-like peptide 1, ghrelin, and leptin.Am J Clin Nutr. 2007; 85: 967-971PubMed Google Scholar In advanced CKD, PEW is a common problem, representing severe and complex processes of muscle loss, poor food intake, inflammation and cardiovascular disease (CVD)1.Fouque D. Kalantar-Zadeh K. Kopple J. et al.A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease.Kidney Int. 2008; 73: 391-398Abstract Full Text Full Text PDF PubMed Scopus (1118) Google Scholar pathways, all of which share intriguing links with the purported ghrelin actions discussed above. Interestingly, the combined effect of ghrelin and higher food intake, but not ghrelin alone, was able to enhance skeletal muscle mitochondrial oxidative capacity and AKT phosphorylation in rats with CKD.30.Barazzoni R. Zhu X. Deboer M. et al.Combined effects of ghrelin and higher food intake enhance skeletal muscle mitochondrial oxidative capacity and AKT phosphorylation in rats with chronic kidney disease.Kidney Int. 2010; 77: 23-28Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar Given the interrelations of PEW with ghrelin, we hypothesized that the implications of low ghrelin in CKD patients are context sensitive and dependent on the presence of PEW. With this purpose, we assessed total ghrelin in a well-characterized cohort of 217 prevalent patients undergoing hemodialysis. The study population consisted of 217 patients undergoing hemodialysis (125 men; 57%) with a median age of 66 (25th–75th percentile 51–74) years. The patients had an average BMI of 24.5±5.2 kg/m2. Of these patients, 55 (25%) had diabetes, 139 (36%) had clinical signs or history of CVD, and 102 (47%) were wasted (subjective global assessment (SGA) >1). Patients underwent hemodialysis three times weekly (4 to 5 h per session) using bicarbonate dialysate. They had undergone hemodialysis for a median period of 29 months (15–58) months and the majority was anuric. Most patients used polyamide membranes (59%), followed by polysulfone (35%). Regarding vascular access, 58% had an arteriovenous fistula, whereas 22 and 20% had grafts and central dialysis catheters, respectively. The general characteristics of the patients according to ghrelin thirds (low third vs the other two-thirds combined) are summarized in Table 1. We should remind the reader that, for a correct interpretation of our results, our definitions of low and high ghrelin correspond to the patients' range. Patients with low ghrelin levels were older, had higher BMI, higher plasma levels of leptin, lower plasma levels of adiponectin, and tended to be more often males. Table 1 also shows the univariate associations between ghrelin levels and selected variables as assessed by Spearman's rank test. Ghrelin concentration positively associated with adiponectin, whereas negatively associated with age, male sex, BMI, and leptin (as well as the leptin/BMI ratio).Table 1General characteristics according to ghrelin thirds and univariate associations with serum ghrelin concentration in 217 hemodialysis patientsaThe low ghrelin group was defined as ghrelin values below the 33rd percentile (lower third) of distribution.Low ghrelin (n=72)High ghrelin (n=145)P-valuebSignificantly different from the low ghrelin group if P<0.05, as assessed by Mann–Whitney U test or χ2 test.ρcUnivariate correlation with ghrelin concentration as assessed by Spearman's rank test; asterisks denote statistical significance as follows: **P<0.01; ***P<0.001.Ghrelin, pg/ml231 (173–261)423 (367–561)——Age, years69 (55–80)dMedian value; 25th to 75th percentile shown in parentheses (all such values).63 (50–72)0.006−0.17**Men, %67ePrevalence, shown in percentage (all such values).530.05—Dialysis vintage, months30 (15–55)28 (14–58)0.9−0.02Diabetes, %29.223.40.3—CVD, %66.762.80.6—PEWfPEW was defined as Subjective Global Assessment >1., %44.448.30.6—BMI, kg/m225.6±4.8gAverage±s.d. (all such values).24.0±5.30.01−0.26***Total cholesterol, mmol/l4.3±1.14.4±1.00.70.09Serum albumin, g/l34.2±4.634.9±4.40.40.02CRP, mg/l6.5 (2.9–17.0)7.0 (2.5–22.5)0.90.07Nt-Pro-BNP, pg/l8.4 (3.3–21.7)7.3 (2.8–33.9)0.80.02Adiponectin, μg/ml19.2 (11.9–26.2)23.9 (15.4–32.7)0.0010.34***Leptin, ng/ml19.9 (8.4–64.4)13.4 (5.1–44.2)0.01−0.23***Leptin/BMI0.81 (0.40–2.22)0.58 (0.24–1.90)0.02−0.20**Abbreviations: BMI, body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; Nt-Pro-BNP, N-terminal prohormone brain natriuretic peptide; PEW, protein-energy wasting.a The low ghrelin group was defined as ghrelin values below the 33rd percentile (lower third) of distribution.b Significantly different from the low ghrelin group if P<0.05, as assessed by Mann–Whitney U test or χ2 test.c Univariate correlation with ghrelin concentration as assessed by Spearman's rank test; asterisks denote statistical significance as follows: **P<0.01; ***P<0.001.d Median value; 25th to 75th percentile shown in parentheses (all such values).e Prevalence, shown in percentage (all such values).f PEW was defined as Subjective Global Assessment >1.g Average±s.d. (all such values). Open table in a new tab Abbreviations: BMI, body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; Nt-Pro-BNP, N-terminal prohormone brain natriuretic peptide; PEW, protein-energy wasting. Survival analysis was determined after a median follow-up period of 31 (20–38) months. During this period, 83 (38%) deaths occurred, of which 36 (44% of all deaths) were because of purportedly CVD-related causes. The impact of ghrelin levels on outcome was studied by the Kaplan–Meier method using the high ghrelin group (middle and high thirds combined) as the reference. Patients with low ghrelin levels had a worse all-cause mortality (log-rank (χ2) 5.50; P=0.01). Crude and adjusted Cox proportional hazard ratios (HRs) for mortality showed that patients with low ghrelin values had a significant crude HR (compared with patients with high ghrelin) of 1.68 (95% confidence interval (CI) 1.08–2.60) that persisted after adjustment of age, sex, and dialysis vintage (HR 1.55, 95% CI 0.99–2.40), but disappeared after further adjustment for comorbidities. We then studied the implications of low ghrelin levels in the context of PEW. The clinical and biochemical characteristics of the patients according to this categorization are detailed in Table 2. Patients with low ghrelin were older, had higher BMI, and presented lower adiponectin. Patients with PEW were also older and exhibited signs of inflammation (increased C-reactive protein (CRP) levels and lower serum albumin) and elevated N-terminal prohormone brain natriuretic peptide values. A significant ghrelin×PEW interaction was found for CRP and leptin values: patients with both PEW and low ghrelin values exhibited the highest CRP concentrations and the highest leptin values (Figure 1). The latter was true even after normalization by BMI, that is, leptin levels indexed to BMI.Table 2Clinical and biochemical characteristics in 217 hemodialysis patients, according to ghrelin and nutritional statusaThe low ghrelin group was defined as ghrelin values below the 33rd percentile (lower third) of distribution, whereas PEW was defined as Subjective Global Assessment >1.Not wasted (n=115)Wasted (n=102)High ghrelin (n=75)Low ghrelin (n=40)High ghrelin (n=70)Low ghrelin (n=32)MANOVAbTwo-factor MANOVA. Significant (P<0.05) effects are given for ghrelin (G), PEW (W), and the interaction ghrelin×PEW (G×W).Ghrelin, pg/ml423 (358–522)cMedian value; 25th to 75th percentile shown in parentheses (all such values).219.2 (138–253)435 (382–619)256 (212–279)—Age, years64 (48–73)66 (46–79)62 (53–71)75 (65–81)G, WMen, %47dPrevalence, shown in percentage (all such values).8046500.004eAssessed by χ2 test.Vintage, months28 (13–57)34 (19–59)29 (14–66)22 (11–54)NSDiabetes, %2027.527.131.3NSeAssessed by χ2 test.CVD, %57.362.568.571.8NSeAssessed by χ2 test.BMI, kg/m224.9±5.8fAverage±s.d. (all such values).25.7±3.922.9±4.525.6±5.9GSerum albumin, g/l36.5±3.936.0±3.333.2±4.331.9±5.1WTotal cholesterol, mmol/l4.5±1.14.4±1.04.2±0.94.2±1.2NSCRP, mg/l4.1 (1.8–15.5)5.2 (1.8–8.2)10.0 (3.8–27.0)12.4 (4.1–31)W, G×WNt-Pro-BNP, pg/l4.6 (2.2–12.9)3.7 (1.7–17.1)11.5 (5.2–3.5)14.3 (3.2–3.5)WAdiponectin, μg/ml22.2 (14.6–29.9)17.2 (10.6–25.7)26.9 (15.4–37.9)23.5 (14.3–26.7)G, WLeptin, ng/ml20.4 (6.0–61.4)17.4 (7.8–44.7)8.0 (3.6–22.5)34.6 (8.9–91.3)G×WLeptin/BMI0.78 (0.27–2.16)0.74 (0.34–1.65)0.37 (0.20–0.85)1.25 (0.50–3.35)G×WAbbreviations: BMI, body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; MANOVA, multivariable analysis of variance; NS, nonsignificant; Nt-Pro-BNP, N-terminal prohormone brain natriuretic peptide; PEW, protein-energy wasting.a The low ghrelin group was defined as ghrelin values below the 33rd percentile (lower third) of distribution, whereas PEW was defined as Subjective Global Assessment >1.b Two-factor MANOVA. Significant (P<0.05) effects are given for ghrelin (G), PEW (W), and the interaction ghrelin×PEW (G×W).c Median value; 25th to 75th percentile shown in parentheses (all such values).d Prevalence, shown in percentage (all such values).e Assessed by χ2 test.f Average±s.d. (all such values). Open table in a new tab Abbreviations: BMI, body mass index; CRP, C-reactive protein; CVD, cardiovascular disease; MANOVA, multivariable analysis of variance; NS, nonsignificant; Nt-Pro-BNP, N-terminal prohormone brain natriuretic peptide; PEW, protein-energy wasting. Survival analysis for these four groups showed, as expected, a negative impact of PEW but also a detrimental impact for concurrent low ghrelin values. Thus, across the four ghrelin-PEW categories, the percentage of deaths during follow-up (both from all- and CVD-related causes) was incrementally higher (Table 3). In Kaplan–Meier curves, this group division resulted in a worse outcome because of both all-cause (log-rank (χ2) 24.61; P<0.0001) and CVD-related (log-rank (χ2) 15.55; P=0.001) mortality (Figure 2). Crude and adjusted Cox proportional HRs are depicted in Table 3, choosing as the reference the group without PEW and elevated ghrelin values. Regarding all-cause mortality, both wasted groups exhibited a worse outcome in crude and adjusted analysis, being considerably worse in magnitude for the group of wasted patients who also had low ghrelin. The HR of wasted patients with low ghrelin compared with wasted patients with high ghrelin was 2.05 (95% CI 1.17–3.57), a difference that persisted after multivariate adjustment (HR 1.96, 95% CI 1.11–3.50). Regarding CVD-related mortality, it was only the group of patients with both PEW and low ghrelin values who had worse outcome in both crude and adjusted models (Table 3). The HR for CVD mortality of wasted patients with low ghrelin compared with wasted patients with high ghrelin was 2.63 (95% CI 1.15–5.99), a difference that persisted after multivariate adjustment (HR 2.78, 95% CI 1.18–6.52). As a sensitivity analysis, results were confirmed by excluding patients within the middle third of ghrelin, and comparing bottom vs top third only; HRs continued being statistically significant in both uni- and multi-variate models (data not shown). The causes of death are detailed in Table 4.Table 3Crude and adjusted all-cause and CVD-related mortality according to ghrelin and wasting groupsaIndicated are univariate and multivariate HRs and 95% CIs for all-cause and CVD mortality. The group of patients who were non-wasted (defined as Subjective Global Assessment >1) and had high ghrelin (grouping middle and high thirds of serum ghrelin together) was used as a reference.Non-wasted, low ghrelinWasted, high ghrelinWasted, low ghrelinModelCovariatesHR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-valueAll-cause mortality 1Crude1.51 (0.72–3.06)0.22.09 (1.17–3.74)0.014.33 (2.30–8.16)<0.0001 21+age, sex, and vintagebAge was categorized according to <45, 45–65, and >65 years, selecting the youngest group as the reference; dialysis vintage was dichotomized to the round approximation of the median value (2 years), using the shorter vintage as the reference category; female sex, the absence of diabetes, and of CVD were considered reference categories.1.55 (0.74–3.18)0.22.37 (1.31–4.32)0.0043.53 (1.84–6.74)0.0002 32+diabetes and CVD1.37 (0.64–2.82)0.42.24 (1.24–4.16)0.0073.34 (1.74–6.41)0.0003Cardiovascular-related mortality 1Crude0.92 (0.25–2.85)0.91.61 (0.68–3.94)0.34.36 (1.79–10.86)0.001 21+age, sex, and vintagebAge was categorized according to <45, 45–65, and >65 years, selecting the youngest group as the reference; dialysis vintage was dichotomized to the round approximation of the median value (2 years), using the shorter vintage as the reference category; female sex, the absence of diabetes, and of CVD were considered reference categories.0.94 (0.25–2.92)0.91.77 (0.73–4.44)0.23.88 (1.56–9.50)0.003 32+diabetes and CVD0.78 (0.21–2.44)0.71.64 (0.68–4.08)0.33.54 (1.40–8.91)0.008Abbreviations: CI, confidence interval; CVD, cardiovascular disease; HR, hazard ratio.a Indicated are univariate and multivariate HRs and 95% CIs for all-cause and CVD mortality. The group of patients who were non-wasted (defined as Subjective Global Assessment >1) and had high ghrelin (grouping middle and high thirds of serum ghrelin together) was used as a reference.b Age was categorized according to <45, 45–65, and >65 years, selecting the youngest group as the reference; dialysis vintage was dichotomized to the round approximation of the median value (2 years), using the shorter vintage as the reference category; female sex, the absence of diabetes, and of CVD were considered reference categories. Open table in a new tab Table 4Individual causes of death according to ghrelin and nutritional statusaIndicated are the causes and number of deaths (n) in each category.Not wastedWastedHigh ghrelin (n=75)Low ghrelin (n=40)High ghrelin (n=70)Low ghrelin (n=32)Cardiac arrest/sudden death4122Myocardial infarction4—36Cerebrovascular accident——31Other causes of cardiac death1332Hemorrhage1—1—Pulmonary edema——1—Malignancy11—2Infection/septicemia1372ESRD treatment withdrawn3221Patient refused further RRT——2—Cachexia1—2—Other——21Uncertain/not determined2334All deathsbIndicated are the number of deaths and percentage, expressed as a proportion of the total number of patients in the group. The proportion of deaths was incrementally higher across the group as assessed by χ2 test (P=0.004 for all deaths and P=0.03 for CVD deaths)., n (%)18 (24%)13 (32%)31 (44%)21 (65%)CVD deathsbIndicated are the number of deaths and percentage, expressed as a proportion of the total number of patients in the group. The proportion of deaths was incrementally higher across the group as assessed by χ2 test (P=0.004 for all deaths and P=0.03 for CVD deaths)., n (%)9 (12%)4 (10%)11 (17%)11 (35%)Abbreviations: CVD, cardiovascular disease; ESRD, end-stage renal disease; RRT; renal replacement therapy.a Indicated are the causes and number of deaths (n) in each category.b Indicated are the number of deaths and percentage, expressed as a proportion of the total number of patients in the group. The proportion of deaths was incrementally higher across the group as assessed by χ2 test (P=0.004 for all deaths and P=0.03 for CVD deaths). Open table in a new tab Abbreviations: CI, confidence interval; CVD, cardiovascular disease; HR, hazard ratio. Abbreviations: CVD, cardiovascular disease; ESRD, end-stage renal disease; RRT; renal replacement therapy. This study reports, for the first time in CKD and we believe in any patient group, an increased mortality risk for patients with low ghrelin values. Although this effect was modest and did not stand full adjustment for confounders, the prognostic value of low ghrelin on outcome seemed magnified in the context of PEW. Thus, wasted patients with low ghrelin concentration presented the highest mortality risk, especially cardiovascular-related mortality. This group of patients showed at the same time abnormally elevated serum CRP and leptin values, forming altogether a pattern of concomitant conditions that seem to fit with the purported anti-inflammatory,19.Li W.G. Gavrila D. Liu X. et al.Ghrelin inhibits proinflammatory responses and nuclear factor-kappaB activation in human endothelial cells.Circulation. 2004; 109: 2221-2226Crossref PubMed Scopus (441) Google Scholar,20.Dixit V.D. Schaffer E.M. Pyle R.S. et al.Ghrelin inhibits leptin- and activation-induced proinflammatory cytokine expression by human monocytes and T cells.J Clin Invest. 2004; 114: 57-66Crossref PubMed Scopus (703) Google Scholar cardioprotective,16.Nagaya N. Moriya J. Yasumura Y. et al.Effects of ghrelin administration on left ventricular function, exercise capacity, and muscle wasting in patients with chronic heart failure.Circulation. 2004; 110: 3674-3679Crossref PubMed Scopus (417) Google Scholar,17.Nagaya N. Uematsu M. Kojima M. et al.Chronic administration of ghrelin improves left ventricular dysfunction and attenuates development of cardiac cachexia in rats with heart failure.Circulation. 2001; 104: 1430-1435Crossref PubMed Scopus (437) Google Scholar and orexigenic14.Ashby D.R. Ford H.E. Wynne K.J. et al.Sustained appetite improvement in malnourished dialysis patients by daily ghrelin treatment.Kidney Int. 2009; 76: 199-206Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar effects attributed to ghrelin administration. Our study therefore identifies a group of dialysis patients who could, at least in theory, benefit from ghrelin treatment.13.Wynne K. Giannitsopoulou K. Small C.J. et al.Subcutaneous ghrelin enhances acute food intake in malnourished patients who receive maintenance peritoneal dialysis: a randomized,