HDL-inflammatory index correlates with poor outcome in hemodialysis patients
2007; Elsevier BV; Volume: 72; Issue: 9 Linguagem: Inglês
10.1038/sj.ki.5002491
ISSN1523-1755
AutoresKamyar Kalantar‐Zadeh, Joel D. Kopple, Naeimeh Kamranpour, Alan M. Fogelman, Mohamad Navab,
Tópico(s)Inflammatory Biomarkers in Disease Prognosis
ResumoOxidative stress and cardiovascular disease are risk factor of patients with chronic kidney disease (CKD) on maintenance hemodialysis. We used the fluorescence of low-density lipoprotein as an index of its proinflammatory potential to examine any role that high-density lipoprotein (HDL) might have in promoting this effect. The total body fat of the patients was measured by means of near-infrared interactance and their quality of life by means of SF36 questionnaires. In 189 randomly selected patients, followed for 30 months, HDL was found to be significantly anti-inflammatory but with a large standard deviation. Fully 17% of the patients had a decidedly proinflammatory index along with inferior SF36 scores. The patients were divided into 10% increments of total body fat percentages up to 40%. HDL was found to be progressively proinflammatory the higher the body fat content. Patients with a higher HDL proinflammatory index had a higher 30-month adjusted hazard ratio for death than those whose HDL were seen to be anti-inflammatory. Our findings suggest an important role of inflammatory HDL in patients with CKD leading to poor outcome. Oxidative stress and cardiovascular disease are risk factor of patients with chronic kidney disease (CKD) on maintenance hemodialysis. We used the fluorescence of low-density lipoprotein as an index of its proinflammatory potential to examine any role that high-density lipoprotein (HDL) might have in promoting this effect. The total body fat of the patients was measured by means of near-infrared interactance and their quality of life by means of SF36 questionnaires. In 189 randomly selected patients, followed for 30 months, HDL was found to be significantly anti-inflammatory but with a large standard deviation. Fully 17% of the patients had a decidedly proinflammatory index along with inferior SF36 scores. The patients were divided into 10% increments of total body fat percentages up to 40%. HDL was found to be progressively proinflammatory the higher the body fat content. Patients with a higher HDL proinflammatory index had a higher 30-month adjusted hazard ratio for death than those whose HDL were seen to be anti-inflammatory. Our findings suggest an important role of inflammatory HDL in patients with CKD leading to poor outcome. Individuals with chronic kidney disease (CKD) have a very high mortality rate and a high burden of cardiovascular disease.1.Go A.S. Chertow G.M. Fan D. et al.Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.N Engl J Med. 2004; 351: 1296-1305Crossref PubMed Scopus (8981) Google Scholar At least one of every five CKD patients requiring maintenance dialysis treatment, currently ∼400 000 individuals in the United States, dies every year.2.National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases United States Renal Data System: Excerpts from the USRDS 2005 Annual Data Report: Atlas of End-Stage Renal Disease in the United States.Am J Kidney Dis. 2006; 47: 1-286PubMed Google Scholar This results in a 5-year survival rate of 1.0. Table 1 shows the relevant demographic, clinical, and laboratory measures in the two categories of MHD patients (HII<1.0 versus ≥1.0). The MHD patients with HII≥1 had significantly higher comorbidities according to the Charlson score and worse self-reported quality of life, but higher body mass index, larger total body fat, and thicker upper arm skinfolds. Among laboratory measures, serum albumin levels, one of the strongest predictors of survival in dialysis patients,19.Kalantar-Zadeh K. Kilpatrick R.D. Kuwae N. et al.Revisiting mortality predictability of serum albumin in the dialysis population: time dependency, longitudinal changes and population-attributable fraction.Nephrol Dial Transplant. 2005; 20: 1880-1889Crossref PubMed Scopus (293) Google Scholar were not significantly different between the two HII groups. Similarly, serum lipoprotein levels, including HDL-cholesterol, were similar in the two groups, suggesting that HII is independent of the serum HDL-cholesterol level. Among inflammatory markers, serum C-reactive protein (CRP) was slightly higher in the HII≥1 group, but interleukin (IL)-6 and tumor necrosis factor-α were similar in the two groups of patients. Predialysis serum creatinine,20.Kopple J.D. Greene T. Chumlea W.C. et al.Relationship between nutritional status and the glomerular filtration rate: results from the MDRD study.Kidney Int. 2000; 57: 1688-1703Abstract Full Text Full Text PDF PubMed Scopus (334) Google Scholar an indicator of muscle mass and probably meat intake, was lower in subjects with HII≥1.Table 1Relevant demographic, clinical, and laboratory values according to the HII cutoff level of 1.0 in 189 MHD patients at baselineHII<1HII≥1P-valuesNumber of MHD patients156 (83%)33 (17%)Gender (% women)42610.05Diabetes mellitus (%)50670.08Race (% Blacks)28230.3Ethnicity (% Hispanics)51580.3Age (years)54±156±20.5Dialysis vintage (months)38±3432±230.4Charlson comorbidity score1.9±1.42.7±1.50.003SF36 quality of life score57±2647±170.02Body composition NIR measured body fat (%)24.4±10.631.6±9.2<0.001 Body mass index (kg/m2)25.2±5.528.1±7.90.02 Triceps skinfold (mm)15.9±10.020.5±10.50.02 Biceps skinfold (mm)10.0±7.913.1±8.60.05Dialysis dose (Kt/V)1.60±0.271.72±0.280.02Protein intake (g/kg/day)1.07±0.241.09±0.230.6Serum concentrations HII0.297±0.3281.455±0.524<0.001 Albumin (g/dl)3.92±0.393.83±0.350.17 Ferritin (ng/ml)634±389777±6970.05 Iron saturation (%)34±1240±150.01 Creatinine (mg/dl)11.3±3.38.9±2.8<0.001 Total cholesterol (mg/dl)147±39146±280.9 LDL (mg/dl)81±3177±200.5 HDL (mg/dl)38±1436±160.5 Triglyceride (mg/dl)129±113131±550.9 Total homocysteine (μmol/l)27±1225±120.4 CRP (mg/l)5.2±6.76.7±6.20.03*P-values for CRP, IL-6, and TNF-α are based on the logarithmic values of these measures. IL-6 (pg/ml)18±6113±270.7*P-values for CRP, IL-6, and TNF-α are based on the logarithmic values of these measures. TNF-α (pg/ml)9.3±11.97.1±4.70.3*P-values for CRP, IL-6, and TNF-α are based on the logarithmic values of these measures.CRP, C-reactive protein; HDL, high-density lipoprotein; HII, HDL-inflammatory index; IL-6, interleukin-6; LDL, low-density lipoprotein; MHD, maintenance hemodialysis; NIR, near infrared; SF36, short form quality of life score with 36 questions; TNF, tumor necrosis factor.P-value pertains to t-test.* P-values for CRP, IL-6, and TNF-α are based on the logarithmic values of these measures. Open table in a new tab CRP, C-reactive protein; HDL, high-density lipoprotein; HII, HDL-inflammatory index; IL-6, interleukin-6; LDL, low-density lipoprotein; MHD, maintenance hemodialysis; NIR, near infrared; SF36, short form quality of life score with 36 questions; TNF, tumor necrosis factor. P-value pertains to t-test. Table 2 shows the association between SF36 scores and HII. All eight scales and the two main dimensions (physical and mental health) of the SF36 had inverse correlations with HII, that is, better self-reported quality of life with lower HII values, although the correlation coefficients were small. The reported quality of life was consistently worse among those who had HII≥1; the difference between the two HII groups was more prominent for the scales Physical Function and Role Emotional.Table 2SF36 quality of life score and its association with HIICorrelation with HII levelsHII<1 (n=156)HII≥1 (n=33)P-valuesSF36 total score-0.28 (P<0.001)57±2647±170.02SF36 dimensions (2) SF-36 mental health-0.26 (P<0.001)59±2550±160.03 SF-36 physical health-0.24 (P<0.001)51±2741±180.03SF36 scales (8) Body pain-0.14 (P=0.06)59±2955±260.3 General health-0.18 (P=0.02)44±2240±210.16 Mental health-0.11 (P=0.16)67±2163±180.17 Physical function-0.28 (P<0.001)54±3338±280.008 Role emotional-0.20 (P=0.01)64±8438±440.04 Role physical-0.21 (P=0.006)49±8424±370.06 Functionality-0.16 (P=0.04)69±3167±230.4 Vitality-0.14 (P=0.07)49±2442±210.06HII, high-density lipoprotein inflammatory index; SF36, short form quality of life score with 36 questions.The first column shows the correlation coefficients between the two main dimensions and eight scales of the health-related quality of life and HII in 189 MHD patients. The following columns compare SF36 scores in patients with HII <1 versus ≥1 (P-value is based on the t-test). Open table in a new tab HII, high-density lipoprotein inflammatory index; SF36, short form quality of life score with 36 questions. The first column shows the correlation coefficients between the two main dimensions and eight scales of the health-related quality of life and HII in 189 MHD patients. The following columns compare SF36 scores in patients with HII <1 versus ≥1 (P-value is based on the t-test). To further examine the nature of the association between the measured body fat and HII, the near infrared (NIR) measured total body fat categories were divided into five a priori selected categories of <10, ≥40 and three 10% groups in-between. As shown in Table 3, both the absolute HII level and the proportion of patients with an HII≥1 were progressively greater in MHD patients with a larger percent of body fat. To examine if the association between the total body fat and HII was due to other confounding demographic or clinical variables, logistic regression analyses were conducted. Table 4 and Figure 2 show the odds ratios of HII≥1 in each of the body fat categories. Comparing the 10–19.9% body fat group, the odds of an HII≥1 was 4–10 times higher in MHD patients with larger total body fat percent, and these associations were independent of other covariates.Table 3The association between total body fat, measured by the NIR interactance, and the serum HII concentrationNIR body fat (%)Number of MHD patients in each categoryHII (mean±s.d.)Number of MHD patients with HII≥1<1014 (7%)0.099±0.2340 (0%)10–19.952 (28%)0.405±0.5153 (6%)20–29.951 (27%)0.531±0.48510 (20%)30–39.951 (27%)0.584±0.61713 (25%)≥4021 (11%)0.738±0.8167 (35%)All patients189 (100%)0.501±0.57433 (17%)HII, high-density lipoprotein inflammatory index; MHD, maintenance hemodialysis; NIR, near infrared.Body fat is categorized into five a priori categories based on increments of 10%. Open table in a new tab Table 4Odds ratio of HII≥1 based on the body fat categories in the cohort in 189 MHD patients of the NIED StudyNIR body fat (%)UnadjustedCase-mix adjustedaCase-mix models are adjusted for age, gender, race and ethnicity, diabetes, Charlson comorbidity scale, and dialysis vintage. Case-mix and MICS-adjusted models are also controlled for serum albumin, CRP, and IL-6.Case-mix and MICS adjustedbNone of the MHD patients with a total body fat <10% had HII≥1; hence, a mathematic OR cannot be calculated, because the logarithm of OR is minus infinity for this group.<10bNone of the MHD patients with a total body fat <10% had HII≥1; hence, a mathematic OR cannot be calculated, because the logarithm of OR is minus infinity for this group.<1.0<1.0<1.010–19.9 (reference group)1.01.01.020–29.94.0 (1.1–15.4)4.7 (1.1–19.5)4.8 (1.2–20.0)P=0.04P=0.03P=0.0330–39.95.6 (1.5–21.0)6.5 (1.3–31.8)6.1 (1.2–30.2)P=0.01P=0.02P=0.03≥408.8 (2.0–38.8)10.6 (1.7–64.6)9.7 (1.6–59.4)P=0.004P=0.01P=0.02CRP, C-reactive protein; HII, high-density lipoprotein inflammatory index; IL-6, interleukin-6; MHD, maintenance hemodialysis; MICS, malnutrition–inflammation–cachexia syndrome; NIED, Nutritional and Inflammatory Evaluation in Dialysis Study; NIR, near infrared.a Case-mix models are adjusted for age, gender, race and ethnicity, diabetes, Charlson comorbidity scale, and dialysis vintage. Case-mix and MICS-adjusted models are also controlled for serum albumin, CRP, and IL-6.b None of the MHD patients with a total body fat <10% had HII≥1; hence, a mathematic OR cannot be calculated, because the logarithm of OR is minus infinity for this group. Open table in a new tab HII, high-density lipoprotein inflammatory index; MHD, maintenance hemodialysis; NIR, near infrared. Body fat is categorized into five a priori categories based on increments of 10%. CRP, C-reactive protein; HII, high-density lipoprotein inflammatory index; IL-6, interleukin-6; MHD, maintenance hemodialysis; MICS, malnutrition–inflammation–cachexia syndrome; NIED, Nutritional and Inflammatory Evaluation in Dialysis Study; NIR, near infrared. During the 30 months of prospective longitudinal observation (1 July 2002 through 31 December 2004), 46 patients died, 22 underwent renal transplantation, and 4 were lost to follow-up. The remaining 117 subjects were confirmed alive as of early January 2005. Among patients with HII≥1, 35 subjects (22%) died versus 11 subjects (33%) in those with HII<1. To examine whether HII≥1 at the baseline of the cohort was associated with a higher death rate, Kaplan–Meier plots were constructed. Figure 3 shows that by the end of the 30 months of observation, the MHD patients with an HII≥1 had worse survival, although this difference was not statistically significant. The survival difference between the two HII groups was more prominent and statically significant after adjusting for case-mix covariates (P=0.03; Figure 3, lower panel). Table 5 compares the hazard ratios of HII≥1 with serum albumin<3.8 g/dl, CRP≥10 mg/l, and IL-6≥10 pg/ml. In the fully adjusted model, HII≥1 was associated with 2.47-fold higher risk of death compared to HII<1 (95% confidence interval: 1.14–5.34, P=0.02). Among other nutritional and inflammatory measures, hypoalbuminemia was the only statistically significant predictor of death (Table 5). Inclusion of serum creatinine, a possible muscle mass indicator, in the fully adjusted Cox model resulted in similar death hazard ratio estimates for HII≥1 (data not shown).Table 5Hazard ratio of 30-month death according to the HII cutoff level of 1 in 189 MHD patients of the NIED Study3-year hazard ratio of death in 189 MHD patientsUnadjustedCase-mixaCase-mix model is adjusted for age, race and ethnicity (Blacks, Asians, and Hispanics), diabetes mellitus, Charlson comorbidity score, dialysis vintage, dialysis dose (Kt/V), and body fat percentage according to the NIR measurements.Case-mix and MICS (full model)bCase-mix and MICS model (the full model) also includes serum albumin (<3.8 versus ≥3.8), CRP (≥10 versus less), and IL-6 (≥10 versus less).HII≥11.61 (0.82–3.18)2.37 (1.10–5.12)2.47 (1.14–5.34)P=0.16P=0.03P=0.02Albumin<3.8 g/dl3.41 (1.89–6.19)2.49 (1.26–4.09)2.39 (1.18–4.87)P<0.001P=0.01P=0.02CRP≥10 mg/l1.39 (0.67–2.87)1.02 (0.48–1.23)0.81 (0.37–1.77)P=0.4P=0.9P=0.6IL-6≥10 pg/ml1.76 (1.00–3.14)1.52 (0.81–2.83)1.30 (0.68–2.46)P=0.05P=0.19P=0.4CRP, C-reactive protein; HII, high-density lipoprotein inflammatory index; IL-6, interleukin-6; MHD, maintenance hemodialysis; MICS, malnutrition–inflammation–cachexia syndrome; NIED, Nutritional and Inflammatory Evaluation in Dialysis Study.a Case-mix model is adjusted for age, race and ethnicity (Blacks, Asians, and Hispanics), diabetes mellitus, Charlson comorbidity score, dialysis vintage, dialysis dose (Kt/V), and body fat percentage according to the NIR measurements.b Case-mix and MICS model (the full model) also includes serum albumin (<3.8 versus ≥3.8), CRP (≥10 versus less), and IL-6 (≥10 versus less). Open table in a new tab CRP, C-reactive protein; HII, high-density lipoprotein inflammatory index; IL-6, interleukin-6; MHD, maintenance hemodialysis; MICS, malnutrition–inflammation–cachexia syndrome; NIED, Nutritional and Inflammatory Evaluation in Dialysis Study. Seventeen percent of 189 MHD patients at the start of a 30-month period of observation had an HII≥1. HDL-cholesterol and other lipoprotein levels in the patients with HII≥1 were similar from those with HII<1. All scales of SF36 health-related quality of life were reported worse in subjects with HII≥1. Body composition had a bearing on the HII, in that HII was progressively higher across the larger total body fat categories, and this association was independent of demographic and clinical characteristics or the severity of comorbid conditions. In this prospective study, the MHD patients with HII≥1 had approximately a 2.5-fold higher adjusted death risk independent of other mortality predictors, including body fat or hypoalbuminemia. These findings may have important clinical implications for the management of inflammation and oxidative stress and subsequent atherosclerotic cardiovascular disease in MHD patients. Inflammation and oxidative stress are likely contributors to morbidity, lower health-related quality of life, and mortality in MHD patients.21.Kalantar-Zadeh K. Balakrishnan V.S. The kidney disease wasting: inflammation, oxidative stress, and diet-gene interaction.Hemodial Int. 2006; 10: 315-325Crossref PubMed Scopus (44) Google Scholar Although in the general population higher HDL is associated with better survival, a recent study did not find any association between HDL and survival in MHD patients.22.Kilpatrick R.D. McAllister C.J. Kovesdy C.P. et al.Association between serum lipids and survival in hemodialysis patients and impact of race.J Am Soc Nephrol. 2007; 18: 293-303Crossref PubMed Scopus (195) Google Scholar Qualitative abnormalities of HDL function in MHD patients were reported in an in vitro study by Morena et al.23.Morena M. Cristol J.P. Dantoine T. et al.Protective effects of high-density lipoprotein against oxidative stress are impaired in haemodialysis patients.Nephrol Dial Transplant. 2000; 15: 389-395Crossref PubMed Scopus (66) Google Scholar including the reduction of HDL protective capacity against oxidative stress because of impairment of LDL oxidation prevention and impairment of reverse cholesterol transport by HDL in these patients. Several studies have indicated an association between low serum total cholesterol and poor survival of dialysis patients.22.Kilpatrick R.D. McAllister C.J. Kovesdy C.P. et al.Association between serum lipids and survival in hemodialysis patients and impact of race.J Am Soc Nephrol. 2007; 18: 293-303Crossref PubMed Scopus (195) Google Scholar, 24.Iseki K. Yamazato M. Tozawa M. et al.Hypocholesterolemia is a significant predictor of death in a cohort of chronic hemodialysis patients.Kidney Int. 2002; 61: 1887-1893Abstract Full Text Full Text PDF PubMed Scopus (299) Google Scholar, 25.Lowrie E.G. Lew N.L. Death risk in hemodialysis patients: the predictive value of commonly measured variables and an evaluation of death rate differences between facilities.Am J Kidney Dis. 1990; 15: 458-482Abstract Full Text PDF PubMed Scopus (1757) Google Scholar An observational study by Liu et al.26.Liu Y. Coresh J. Eustace J.A. et al.Association between cholesterol level and mortality in dialysis patients: role of inflammation and malnutrition.JAMA. 2004; 291: 451-459Crossref PubMed Scopus (612) Google Scholar showed that MICS may be the cause of the inverse association between cholesterol and mortality in these patients. In the 4D Study,4.Wanner C. Krane V. Marz W. et al.Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis.N Engl J Med. 2005; 353: 238-248Crossref PubMed Scopus (2206) Google Scholar 1255 diabetic dialysis patients were randomized to receive either atorvastatin (20 mg/day) or placebo for 5 years. The study was recently reported to be negative4.Wanner C. Krane V. Marz W. et al.Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis.N Engl J Med. 2005; 353: 238-248Crossref PubMed Scopus (2206) Google Scholar with a nonsignificant 8% reduction of the primary composite end points achieved by treatment with the cholesterol/LDL-lowering agent. This was in distinct contrast to the recently published CARDS trial (Collaborative Atorvastatin Diabetes Study)27.Colhoun H.M. Betteridge D.J. Durrington P.N. et al.Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial.Lancet. 2004; 364: 685-696Abstract Full Text Full Text PDF PubMed Scopus (3259) Google Scholar in type II diabetic patients who had not yet developed significant kidney disease. Hence, the association between traditional risk factors and survival in CKD patients is likely confounded by factors related to CKD and MHD. Protein-energy malnutrition and inflammation, independently or together as in MICS, are common occurrences in CKD patients.5.Kalantar-Zadeh K. Ikizler T.A. Block G. et al.Malnutrition–inflammation complex syndrome in dialysis patients: causes and consequences.Am J Kidney Dis. 2003; 42: 864-881Abstract Full Text Full Text PDF PubMed Scopus (746) Google Scholar,28.Stenvinkel P. Interactions between inflammation, oxidative stress, and endothelial dysfunction in end-stage renal disease.J Ren Nutr. 2003; 13: 144-148Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar MICS is associated with poor clinical conditions and worse outcomes.29.Stenvinkel P. Malnutrition and chronic inflammation as risk factors for cardiovascular disease in chronic renal failure.Blood Purif. 2001; 19: 143-151Crossref PubMed Scopus (128) Google Scholar,30.Kaysen G.A. Chertow G.M. Adhikarla R. et al.Inflammation and dietary protein intake exert competing effects on serum albumin and creatinine in hemodialysis patients.Kidney Int. 2001; 60: 333-340Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar The confounding effect of MICS on associations between traditional risk factors such as obesity and hypercholesterolemia and clinical outcome is so strong that it even reverses these associations. Hence, a low, rather than a high, body mass index or serum cholesterol level is associated with mortality in MHD patients.24.Iseki K. Yamazato M. Tozawa M. et al.Hypocholesterolemia is a significant predictor of death in a cohort of chronic hemodialysis patients.Kidney Int. 2002; 61: 1887-1893Abstract Full Text Full Text PDF PubMed Scopus (299) Google Scholar,31.Salahudeen A.K. Obesity and
Referência(s)