Contribution of subcutaneous fat accumulation to insulin resistance and atherosclerosis in haemodialysis patients

Nephrology Dialysis Transplantation, Nov 2009

Background. Whereas visceral fat accumulation (VFA) is related to insulin resistance and atherosclerosis in both haemodialysis (HD) patients and the general population, little is known about the role of subcutaneous fat accumulation (SFA). The purpose of the present study was to examine and confirm the relationship between abdominal fat accumulation (AFA) and various clinical parameters in HD patients. Methods. Two hundred and thirty-three HD patients were recruited, including 120 with type 2 diabetes. Abdominal fat distribution was evaluated by computed tomography (CT) scans. Systemic atherosclerosis was assessed by intima–media thickness (IMT) using high-resolution B-mode ultrasonography. The insulin resistance was estimated by the homeostasis model assessment-insulin resistance (HOMA-IR). Results. Spearman's analysis revealed that both VFA and SFA showed a significant relationship with HOMA-IR and also that SFA was correlated significantly with IMT in all HD patients. SFA was an independent risk factor associated with HOMA-IR and IMT in multiple regression analysis. Neither body mass index (BMI) nor VFA was a predominant determinant of HOMA-IR and IMT. IMT in HD patients with high SFA/low BMI groups was significantly higher than in the low SFA/high BMI groups. Conclusion. It appears that there is a close relationship between SFA and insulin resistance or atherosclerosis in HD patients. It was suggested that SFA plays important roles related to carbohydrate or lipid metabolism in HD patients.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://ndt.oxfordjournals.org/content/24/11/3474.full.pdf

Contribution of subcutaneous fat accumulation to insulin resistance and atherosclerosis in haemodialysis patients

Hiromichi Gotoh 0 1 Tomohito Gohda 1 Mitsuo Tanimoto 1 Yoshikazu Gotoh 0 Satoshi Horikoshi 1 Yasuhiko Tomino 1 0 Saiyu Souka Hospital , Japan 1 Division of Nephrology, Department of Internal Medicine, Juntendo University School of Medicine of atherosclerosis and cardiovascular complications. The Background. Whereas visceral fat accumulation (VFA) is causes of atherosclerosis include multiple factors such as related to insulin resistance and atherosclerosis in both obesity, age, smoking, hypertension, hypotension, abnorhaemodialysis (HD) patients and the general population, malities of bone and mineral homeostasis and disturbances little is known about the role of subcutaneous fat accu- of lipid metabolism and insulin resistance. mulation (SFA). The purpose of the present study was to Several methods can be employed to examine the proexamine and confirm the relationship between abdominal gression of atherosclerosis. Quantitative assessment of fat accumulation (AFA) and various clinical parameters in carotid intima-media thickness (IMT) is now regarded as HD patients. a major indicator of atherosclerosis [1] and also an imMethods. Two hundred and thirty-three HD patients were portant marker for cardiovascular morbidity and mortality recruited, including 120 with type 2 diabetes. Abdomi- [2]. More than 20 years ago, Lindner et al. [3] reported nal fat distribution was evaluated by computed tomog- a high incidence of coronary deaths in patients receiving raphy (CT) scans. Systemic atherosclerosis was assessed haemodialysis (HD). In the 1990s, Kawaguchi et al. [4] by intima-media thickness (IMT) using high-resolution B- reported that patients with chronic kidney disease (CKD) mode ultrasonography. The insulin resistance was estimated stage 5 had a much higher prevalence of atherosclerosis by the homeostasis model assessment-insulin resistance than the general population. Recently, Krasniak et al. [5] (HOMA-IR). showed an association between carotid atherosclerosis and Results. Spearman's analysis revealed that both VFA and coronary artery calcification in maintenance HD patients. SFA showed a significant relationship with HOMA-IR and In contrast to the general population, atherosclerosis ofalso that SFA was correlated significantly with IMT in all ten coexists with malnutrition in HD patients. Numerous HD patients. SFA was an independent risk factor associated studies showed that malnutrition is a serious risk factor for with HOMA-IR and IMT in multiple regression analysis. morbidity and mortality in CKD patients not yet on HD [6] Neither body mass index (BMI) nor VFA was a predominant or HD patients [7-11]. Although, body fat mass increases in determinant of HOMA-IR and IMT. IMT in HD patients the first year of maintenance HD [12], 50-68% of HD pawith high SFA/low BMI groups was significantly higher tients older than 65 years were reported to be malnourished than in the low SFA/high BMI groups. based on various clinical parameters [13]. Actually, these Conclusion. It appears that there is a close relationship findings seem to correspond to our clinical understanding between SFA and insulin resistance or atherosclerosis in of HD patients. Since it is possible to measure fat accuHD patients. It was suggested that SFA plays important mulation directly using computed tomography (CT) scans, roles related to carbohydrate or lipid metabolism in HD many studies have demonstrated the association between patients. visceral fat accumulation (VFA) and insulin resistance or systemic atherosclerosis in the general population [14-16]. - However, little is known about the role of SFA in HD patients, and these interesting reports inspired us to examine the relationship between abdominal fat accumulation (AFA) and various clinical parameters of atherosclerosis such as insulin resistance and IMT in HD patients. The purpose of the present study was to examine whether AFA affects insulin resistance and various clinical parameters of atherosclerosis in HD patients. Subjects and methods Patients Two hundred forty-one patients [122 type 2 diabetes mellitus (T2DM), 119 non-DM; 134 men, 107 women] were undergoing HD at Saiyu Souka Hospital (Souka, Saitama Prefecture, Japan). In this study, informed consent was obtained from 233 HD patients before the study. The study protocol was approved by the Ethics Committee of Saiyu Souka Hospital. This study was carried on 233 patients (132 men, 101 women; age 60 11 years, duration of HD 5 5 years). They included 120 diabetic and 113 non-diabetic HD patients. The diabetic HD patients accounted for 51.5% (120/233) of all HD patients. The diagnosis of diabetes was based on a history of diabetes or the criteria in the Report of Expert Committee on the Diagnosis and Classification of Diabetes Mellitus [22]. The prevalence of diabetes in this study was very high compared with that (31.9%) of the Japan Dialysis Outcomes and Practice Pattern (DOPPS) Study [23]. HD patients were maintained on a regular regimen using bicarbonate dialysate three times a week. Dry weight was determined in each patient to achieve a normotensive, oedema-free status based on inferior vena cava diameter, plasma concentration of atrial natriuretic peptide (ANP) and cardiothoracic ratio (CTR). Blood samples for biochemical parameters were drawn during fasting. Measurement of body fat mass Abdominal fat distribution was determined using abdominal CT scans at the level of the umbilicus [2]. CT scanning was performed with the subject in the supine position using a slip ring CT scanner (Siemens Somatom Emotion Duo, Saitama Prefecture, Japan). SFA was clearly defined as the extraperitoneal fat between skin and muscle. Intra-abdominal tissue with the same density as SFA was defined as VFA. SFA and VFA were also measured at the level of the umbilicus. Evaluation of carotid artery IMT Carotid artery IMT was measured by high-resolution real-time B mode ultrasonography (G.E Logic, Saitama Prefecture, Japan). Ultrasonography was performed with the subject in the supine position. The carotid arteries were investigated bilaterally in the longitudinal projections. The examination included sections of 23 cm of common carotid artery just below the carotid bulb. IMT was defined by Salonen et al. [24] as the distance between the leading edge of the first echogenic line (lumenintima interface) and the second echogenic line (mediaadventitia interface) of the far wall. Four measurements from both sites were averaged to give the mean of IMT. Biochemical analysis Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Plasma glucose and serum lipid [cholesterol, high-density lipoprotein (HDL)-cholesterol, low-density lipoprotein (LDL)-cholesterol and triglyceride] levels were measured using standard laboratory methods. Levels of remnant-like particle (RLP-cholesterol were measured using immunoadsorption (RemL-C, Kyowa Hakko, Saitama Prefecture, Japan). Levels of C-reactive protein (CRP) were measured by nephelometry, a latex particle-enhanced immunoassay (N Latex CRP II; Dabe Behring, Tokyo, Japan). Plasma insulin levels were also measured by radioimmunoassay (insulin RIA-BEAD; Dinabot, Saitama Prefecture, Japan). In subjects not receiving insulin with fasting plasma glucose (FPG) levels of 170 mg/dL or less, the insulin resistance was estimated by homeostasis model assessment (HOMA). The HOMA insulin resistance index (HOMA-IR) was calculated using the formula: fasting insulin (U/mL) fasting glucose (mg/dL)/405 [25]. HOMA-IR showed a close correlation Group C Group D Group E Group F (n=85) (n=50) (n=31) (n=67) Fig. 2. Age-adjusted average intimamedia thickness (IMT) levels according to mean BMI and SFA. Group C, low SFA (<130 cm2) and low BMI (<21.9 kg/m2); Group D, high SFA (130 cm2) and low BMI; Group E, low SFA and high BMI (21.9 kg/m2); Group F, high SFA and high BMI. Data were expressed as mean SE. with the insulin-sensitivity index by the standard euglycaemic hyperinsulinaemic clamp as reported by Mathew et al. [25] and Emoto et al. [26]. This index can be applied to subjects with renal failure [27]. Statistical analysis Data were expressed as mean standard deviation (SD) or standard error (SE). Spearmans analysis was applied to examine the relationship between SFA or VFA and various other clinical parameters such as systolic blood pressure and serum lipids. Multiple regression analysis was used to determine the contribution of various factors to HOMA-IR or IMT. A P-value <0.05 was considered statistically significant. Model 1 included age, sex, presence of T2DM, SBP, cholesterol, HDL-cholesterol and LDL-cholesterol for HOMA-IR or IMT to evaluate the predominant determinants in classical factors. Model 2 included all clinical factors. Three analyses shown in Figures 13 were performed to clarify the effect of obesity on IMT. Firstly, patients were divided by mean BMI into two groups as follows: A, low BMI (<21.9 kg/m2) and B, high BMI (21.9 kg/m2), as shown in 1. Secondly, patients were divided into four groups by mean BMI and SFA as follows: C, low SFA (<130 cm2) and low BMI (<21.9 kg/m2); D, high SFA (130 cm2) and low BMI; E, low SFA and high BMI (21.9 kg/m2) and F, high SFA and high BMI (Figure 2). Moreover, patients were also divided into four groups by mean BMI and VFA as follows: G, low VFA (<97.5 cm2) and low BMI (<21.9 kg/m2); H, high VFA (97.5 cm2) and low BMI; I, low VFA and high BMI (21.9 kg/m2) and J, high VFA and high BMI (Figure 3). All groups were analysed by ANOVA for an unbalanced design. All statistical calculations were performed using SAS (Version 9.2, SAS Institute, Cary, NC, USA). Group A (n=135) Group B (n=98) Fig. 1. Average intimamedia thickness (IMT) levels according to mean BMI. Group A, low BMI (<21.9 kg/m2); Group B, high BMI (21.9 kg/m2). Data were expressed as mean SE. Group G Group H Group I Group J (n=94) (n=41) (n=39) (n=59) Fig. 3. Age-adjusted average intimamedia thickness (IMT) levels according to mean BMI and VFA. Group G, low VFA (<97.5 cm2) and low BMI (<21.9 kg/m2); Group H, high VFA (97.5 cm2) and low BMI; Group I, low VFA and high BMI (21.9 kg/m2); Group J, high VFA and high BMI. Data were expressed as mean SE. Results Characteristics of haemodialysis (HD) patients Clinical characteristics of HD patients with or without T2DM are summarized in Table 1. There were significant differences in the duration of HD, systolic blood pressure (SBP), RLP-cholesterol, FPG, CRP, SFA, immunoreactive insulin (IRI), homeostasis model assessment-insulin resistance (HOMA-IR) and IMT between diabetic and nondiabetic groups. However, other clinical parameters did not show any significant differences between these groups. Spearmans correlation analysis of clinical parameters for VFA and SFA As shown in Table 2, both VFA and SFA had significant relationships with HDL-cholesterol, triglyceride, FPG, BMI, IRI, CRP and HOMA-IR in all groups (All (DM + nonDM), DM, non-DM). There was a correlation between VFA and sex, but no correlation between SFA and sex. On the other hand, SFA was correlated significantly with IMT in all groups. Multiple regression analysis of clinical parameters for HOMA-IR or IMT Multiple regression analysis was applied to identify which parameter had an independent relationship with HOMA-IR or IMT (Tables 35). In model 1 of all HD patients, LDL-cholesterol and HDLcholesterol (P < 0.05) were the predominant determinants of HOMA-IR. The presence of T2DM and HDL-cholesterol were closely associated with IMT in addition to age. In model 2 of all HD patients, SFA was predominant determinant of HOMA-IR in addition to RLP-cholesterol and CRP. SFA was also a predominant determinant of IMT in addition to age and the presence of T2DM. Moreover, SFA was an independent factor associated with HOMA-IR and IMT in diabetic HD patients (Table 4), while it was an independent factor associated only with IMT in HD patients without T2DM (Table 5). Average IMT levels according to BMI and SFA or VFA As in Figure 1, IMT in Group B (high BMI) was significantly higher than that in Group A (low BMI). In addition, age-adjusted average IMT in Group D (high SFA and low BMI) was significantly higher than that in Group C (low SFA and low BMI) (Group C, 0.86 0.03 mm; Group D, 0.96 0.03 mm; P < 0.05). Interestingly, age-adjusted average IMT in Group D (high SFA and low BMI) was significantly higher than that in Group E (low SFA and high BMI) (Group D, 0.85 0.04 mm; Group E, 1.04 0.03 mm; P < 0.05). Moreover, age-adjusted average IMT did not show a significant difference among four groups (Group GJ) based on VFA and BMI (Figure 3). Discussion This study revealed that SFA was closely correlated with insulin resistance and carotid atherosclerosis in haemodialysis (HD) patients. VFA rather than SFA was suggested to play a greater role than SFA in the disturbances of carbohydrate and lipid metabolism or insulin resistance in many studies observing the general population [14,16,28] and HD patients [17,29]. In non-diabetic HD patients, Odamaki et al. [17] reported that VFA was strongly correlated with serum lipid abnormalities. Moreover, Yamauchi et al. [29] reported that VFA was related to insulin resistance, and VFA might also be correlated with carotid atherosclerosis in non-diabetic HD patients. According to these reports, it is also assumed that VFA rather than SFA has a close relationship with insulin resistance or carbohydrate and lipid metabolism, as well as with atherosclerosis in HD patients without T2DM. However, the results of the present study were not consistent with many previous studies except for a few studies on the general population [19,20]. We reported that SFA was more closely related to insulin resistance than VFA in 80 HD patients including 25 diabetic HD patients [21]. The subjects of this study were more numerous than in our previous study, and the prevalence of diabetic patients was higher compared with that of the Japan Dialysis and Practice Pattern Study (DOPPS) [23]. As in our previous study, this study also revealed a strong relationship between SFA and insulin resistance or CRP in all HD patients. In addition, whole AFA was also correlated with both of them in Spearmans analysis (homeostasis model assessment-insulin resistance index (HOMA-IR): r = 0.39, P < 0.0001; CRP: r = 0.30, P < 0.0001) (data not shown). Axelesson et al. also reported that truncal fat mass, estimated by dual-energy x-ray absorptiometry (DEXA), correlated with increased inflammationory biomarkers such as CRP and interleukin 6 (IL-6) in CKD stage 5 [30]. While VFA was not correlated with insulin resistance in multiple regression analysis, RLP-cholesterol and CRP may exhibit an independent association with insulin resistance in HD patients. In contrast with these findings, Saijo et al. reported an association between CRP and VFA in healthy Characteristics 61 11 132/101 All (DM + non-DM) DM n= 120 Japanese people [31]. They indicated that VFA might be associated with inflammation. Another purpose of the present study was to investigate the relationship between AFA and various clinical parameters of atherosclerosis. Among groups based on SFA and BMI, IMT in Group D (high SFA and low BMI) was significantly higher than that in Group E (low SFA and high BMI). Moreover, IMT did not show a significant difference among four groups (Group GJ) based on VFA and BMI. Our additional analyses suggested that SFA was the predominant determinant of atherosclerosis, whereas previous studies reported an association between high BMI and carotid atherosclerosis in the general population [32,33]. Carotid atherosclerosis is well recognized as the marker of systemic atherosclerosis [34,35]. IMT is related to cardiovascular atherosclerosis in both HD patients and the general population. For example, Krasniak et al. [5] examined the risk factors of atherosclerosis in HD patients without observing insulin resistance and measuring AFA by CT scan. They could not clearly indicate the association between carotid atherosclerosis and classical major risk factors, such as total cholesterol and LDL-cholesterol. In this study, only HDL-cholesterol and total cholesterol were closely related to carotid atherosclerosis. Moreover, IMT was closely associated with age and the presence of T2DM as well as in previous studies in general [34,35]. Non-DM n= 113 P <0.0001; P <0.01; P <0.05; otherwise non-significant. HD, haemodialysis; DM, type 2 diabetes mellitus; SBP, systolic blood pressure; BMI, body mass index; VFA, visceral fat accumulation; SFA, subcutaneous fat accumulation; HDL, high-density lipoprotein; LDL, lowdensity lipoprotein; RLP, remnant-like particle; HOMA-IR, homeostasis model assessment insulin resistance index; CRP, C-reactive protein; IMT, intimamedia thickness. Model 1: analysis of age, sex, SBP, presence of DM, cholesterol, HDLcholesterol and LDL-cholesterol with HOMA-IR or IMT (however, the presence of DM was excluded when HOMA-IR was a dependent variable). Model 2: analysis of all variables with HOMA-IR or IMT. P <0.0001; P <0.01; P <0.05; otherwise non-significant. HD, haemodialysis; SBP, systolic blood pressure; BMI, body mass index; VFA, visceral fat accumulation; SFA, subcutaneous fat accumulation; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RLP, remnant-like particle; HOMA-IR, homeostasis model assessment insulin resistance index; CRP, C-reactive protein; IMT, intimamedia thickness. Model 1: analysis of sge, sex, SBP, cholesterol, HDL-cholesterol and LDL-cholesterol with HOMA-IR or IMT. Model 2: analysis of all factors with HOMA-IR or IMT. Unexpectedly, among other clinical risk factors, duration of HD did not affect atherosclerosis strongly. Some papers [3,4] proposed that atherosclerosis might be accelerated in long-term HD. On the other hand, Shoji et al. [36] proposed the concept that advanced atherosclerosis in HD 0.57 0.89 P <0.0001; P <0.01; P <0.05; otherwise no significant. HD, haemodialysis; SBP, systolic blood pressure; BMI, body mass index; VFA, visceral fat accumulation; SFA, subcutaneous fat accumulation; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RLP, remnant-like particle; HOMA-IR, homeostasis model assessment insulin resistance index; CRP, C-reactive protein; IMT, intimamedia thickness. Model 1: analysis of age, sex, SBP, cholesterol, HDL-cholesterol and LDL-cholesterol with HOMA-IR or IMT. Model 2: analysis of all factors with HOMA-IR or IMT. patients is not due to HD treatment, but to renal failure and/or metabolic abnormalities secondary to renal failure. Both VFA and SFA decreased with ageing (VFA: r = 0.18, P < 0.05; SFA: r = 0.26, P = 0.005) and duration of HD therapy (VFA: r = 0.20, P < 0.05; SFA: r = 0.18, P < 0.05) in HD patients with T2DM (data not shown). Unexpectedly, the decrease rate of VFA and SFA while undergoing HD therapy showed no difference. On the other hand, Li et al. [37] stated that VFA rather than SFA decreased more when patients lost body weight. In support of the present study, it was suggested that SFA plays important and essential roles related to carbohydrate or lipid metabolism in HD patients. In non-HD patients, adipose tissues accumulate mostly in SFA at a young age. It is recognized that fat distribution from SFA into VFA may occur with ageing, differences in gender and race [38]. Miyazaki et al. examined the effect of pioglitazone on circulating adipocytokine and also on abdominal adipose tissue in type 2 diabetic patients [39,40]. Interestingly, they reported that an increase in plasma adiponectin and fat distribution was observed after pioglitazone treatment. In this report, SFA increased, while VFA decreased at the same time. It appears that this fat distribution is explained by PPAR agonist action. The activation of the PPAR increases adiponectin, one of the fat-derived hormones [41]. Furthermore, Yamauchi et al. [42] examined the relationship between adiponectin and insulin resistance or atherosclerosis. Their reports demonstrated that decreased adiponectin is implicated in the development of insulin resistance or atherosclerosis in mouse models. Although plasma adiponectin levels were not measured in this study, it is well-known that plasma adiponectin levels are high in HD patients [43,44]. It is possible that increased adiponectin levels in HD patients contribute to fat distribution from VFA into SFA, and also development of insulin resistance and atherosclerosis. Increased subcutaneous adipose tissue, that acts in the same way as visceral adipose tissue, might have a major effect on the insulin resistance and atherosclerosis in HD patients. This hypothesis seems to be able to explain the results of our study in which both VFA and SFA decreased based on ageing or duration of HD therapy. Some limitations of the present study should be considered. HOMA-IR and IMT are not definitely established methodologies for CKD patients, although previous reports [5,27] suggested HOMA-IR or IMT can be used in CKD patients. In conclusion, this study revealed a close relationship between SFA and insulin resistance or atherosclerosis in HD patients. To our knowledge, no reports on the relationship between SFA and insulin resistance or atherosclerosis in HD patients have appeared. Conflict of interest statement. Part of this study was presented at the general convention of the Japanese Society for Dialysis Therapy (JSDT), Osaka, Japan, and appears in an abstract form, Jpn Soc Dial Ther 2008; Suppl 41; 415. Received for publication: 15.7.08; Accepted in revised form: 23.5.09 Haemodialysis-induced transient CD16+ monocytopenia and cardiovascular outcome Kyrill S. Rogacev, Maren Ziegelin, Christof Ulrich, Sarah Seiler, Matthias Girndt, Danilo Fliser and Gunnar H. Heine Abstract Background. Haemodialysis with bioincompatible membranes led to transient leukocyte activation and intradialytic leukopenia due to endothelial adherence. After the introduction of biocompatible membranes, only CD16+ (i.e. CD14++CD16+ and CD14(+)CD16+) monocytes showed an impressive transient intra-dialytic decrease. Presently, it is unclear whether this CD16+ monocyte drop is detrimental. We investigated whether a prominent intradialytic decrease of CD16+ monocytes predicts future cardiovascular (CV) events. Methods. We measured leukocyte and monocyte subpopulations in 70 patients before and 10 min after haemodialysis initiation. Patients were stratified by their intra-dialytic CD14++CD16+ monocyte drop (pre-defined major drop: decline of cell counts at 10 min to <50% of pre-dialytic values; pre-defined minor drop: decline to values >50% of pre-dialytic counts). Patients were followed up for 42 2 months; endpoints were CV events and death. Results. Patients with a minor CD14++CD16+ monocyte drop had more CV events than patients with a major drop. In multivariate analysis, a minor CD14++CD16+ monocyte drop was the strongest independent predictor of future CV events [hazard ratio 2.405 (95% CI 1.1924.854)]. Conclusions. These data refute the assumption that a prominent intra-dialytic decrease of CD14++CD16+ monocytes is detrimental. Instead, a minor cell drop could mirror CD14++CD16+ monocyte dysfunction, with inadequate migratory reaction towards an immunologic stimulus posed by membrane and tubing contact. Persistent microinflammation and alterations in immune function, which are characterized by an increase in proinflammatory cytokines such as TNF- and IL-6 [1,2], a shift in monocyte subsets [3] and monocytic dysfunction [4], are thought to contribute to the devastating prognosis of haemodialysis patients [5].


This is a preview of a remote PDF: https://ndt.oxfordjournals.org/content/24/11/3474.full.pdf

Hiromichi Gotoh, Tomohito Gohda, Mitsuo Tanimoto, Yoshikazu Gotoh, Satoshi Horikoshi, Yasuhiko Tomino. Contribution of subcutaneous fat accumulation to insulin resistance and atherosclerosis in haemodialysis patients, Nephrology Dialysis Transplantation, 2009, 3474-3480, DOI: 10.1093/ndt/gfp290