Abstract
Background and objectives
Insulin resistance has been associated with cardiovascular disease in peritoneal dialysis patients. Few studies have addressed the impact of fast transport status or dialysis prescription on insulin resistance. The aim of this study was to test whether insulin resistance is associated with obesity parameters, peritoneal transport rate, and glucose absorption.
Design, setting, participants, & measurements
Insulin resistance was evaluated with homeostasis model assessment method (HOMA-IR), additionally corrected by adiponectin (HOMA-AD). Enrolled patients were prevalent nondiabetics attending at Santo António Hospital Peritoneal Dialysis Unit, who were free of hospitalization or infectious events in the previous 3 months (51 patients aged 50.4±15.9 years, 59% women). Leptin, adiponectin, insulin-like growth factor-binding protein 1 (IGFBP-1), and daily glucose absorption were also measured. Lean tissue index, fat tissue index (FTI), and relative fat mass (rel.FM) were assessed using multifrequency bioimpedance. Patients were categorized according to dialysate to plasma creatinine ratio at 4 hours, 3.86% peritoneal equilibration test, and obesity parameters.
Results
Obesity was present in 49% of patients according to rel.FM. HOMA-IR correlated better with FTI than with body mass index. Significant correlations were found in obese, but not in nonobese patients, between HOMA-IR and leptin, leptin/adiponectin ratio (LAR), and IGFBP-1. HOMA-IR correlated with HOMA-AD, but did not correlate with glucose absorption or transport rate. There were no significant differences in insulin resistance indices, glucose absorption, and body composition parameters between fast and nonfast transporters. A total of 18 patients (35.3%) who had insulin resistance presented with higher LAR and rel.FM (7.3 [12.3, interquartile range] versus 0.7 [1.4, interquartile range], P<0.001, and 39.4±10.1% versus 27.2±11.5%, P=0.002, respectively), lower IGFBP-1 (8.2±7.2 versus 21.0±16.3 ng/ml, P=0.002), but similar glucose absorption and small-solute transport compared with patients without insulin resistance. FTI and LAR were independent correlates of HOMA-IR in multivariate analysis adjusted for glucose absorption and small-solute transport (r=0.82, P<0.001).
Conclusions
Insulin resistance in nondiabetic peritoneal dialysis patients is associated with obesity and LAR independent of glucose absorption and small-solute transport status. Fast transport status was not associated with higher likelihood of obesity or insulin resistance.
Keywords: nutrition, insulin resistance, peritoneal glucose absorption, Adiponectin, Body Mass Index, diabetes mellitus, Insulin-Like Growth Factor Binding Protein 1 Leptin obesity peritoneal dialysis
Introduction
Insulin resistance is already present in early stages of CKD (1), with likely contributions from uremic toxins, inflammation, malnutrition, metabolic acidosis, and vitamin D deficiency leading to acquired defects in the insulin receptor signaling pathway (2).After induction, peritoneal dialysis (PD) has been shown to initially improve insulin resistance in uremic patients, similar to hemodialysis (HD) (3). However, it is feared that with time the cumulative exposure to glucose solutions used in PD might lead to systemic hyperglycemia, obesity, and aggravate insulin resistance, which could contribute to increased cardiovascular risk (4). The incidence of overweight and obese patients is widely prevalent in contemporary PD populations (5,6), although there is much controversy about the major role of peritoneal glucose absorption on body fat accumulation over time (7,8). A single study has evidenced that insulin resistance evaluated by homeostasis model assessment (HOMA-IR) is a predictor of cardiovascular disease in PD patients (9); however, the specific role of peritoneal glucose absorption, small-solute transport, or obesity were not addressed.
Hung et al. (10) have shown that leptin/adiponectin ratio (LAR) and homeostasis model assessment corrected for adiponectin (HOMA-AD) were the better correlates of glucose-disposal rate measured by hyperinsulinemic euglycemic glucose clamp in chronic HD patients. LAR has been recently proposed as a new atherogenic index, since it has been associated with cardiovascular events in PD (6). However, no study has explored the association between these newly adipokine-based insulin resistance indices and the HOMA-IR in PD patients.
Therefore, we aimed to test whether insulin resistance is associated with obesity parameters, peritoneal transport rate, and glucose absorption, by using three insulin resistance indices (HOMA-IR, HOMA-AD, and LAR) and bioimpedance analysis in stable PD patients. We further evaluated insulin-like growth factor-binding protein 1 (IGFBP-1) as a marker of hepatic insulin sensitivity, and explored its association with obesity. Additionally, we evaluated the determinants of insulin resistance measured by HOMA-IR in a group of patients undergoing PD for >1 year.
Materials and Methods
Patients
This cross-sectional study enrolled all the 51 prevalent nondiabetic PD patients currently attending Santo António Hospital Peritoneal Dialysis Unit, who were free of hospitalization or infectious events in the previous 3 months. All patients were treated with low-glucose degradation product solutions; 3.86% glucose exchanges were not routinely used. Icodextrin was used in 21 patients (41.2%). None of the patients were currently taking corticosteroids. All patients provided informed consent and the study was approved by the Centro Hospitalar do Porto Ethics Committee [approval #061/12(039-DEFI/059-CES)].
Laboratory Measurements
Venous blood samples were obtained after an 8–10-hour overnight fasting period, on the morning of the peritoneal equilibration test (PET). The preceding overnight dwell was standardized to 1.36% glucose solution. Blood samples were drawn after the last dwell drainage and before starting the PET. Serum was allowed to clot at room temperature for no longer than 60 minutes and was then separated from cells by centrifugation and divided in two aliquots, one for immediate assay, the other for storage at –70°C until analysis for adiponectin, leptin, and IGFBP-1.
Creatinine was measured by the compensated Jaffe method. The dialysate creatinine concentration was corrected for interference by glucose according to our laboratory standards. Serum albumin was measured by bromocresol green dye (Cobas Integra 800, Roche Diagnostics GmbH). Insulin was measured by a sandwich assay on an electrochemiluminescence immunoassay analyzer (Cobas E 170, Roche Diagnostics GmbH). Glucose and C-reactive protein was measured using the Roche methods in a Roche autoanalyzer (COBAS Integra 800, Roche Diagnostics GmbH). Serum leptin, adiponectin, and IGFBP-1, were measured using ELISA (Mediagnost, Reutlingen, Germany).
Derived Insulin Resistance Indices
Insulin resistance was assessed using HOMA-IR, HOMA-AD, and LAR, according to the following equations (10):
 |
 |
 |
Patients with HOMA-IR ≥2.2 were considered to have insulin resistance, as this value corresponds to the HOMA-IR 50th percentile in the larger validation study about HOMA-IR against the euglycemic, hyperinsulinemic clamp in nondiabetic patients (11).
Dialysis Adequacy, Urea Kinetics, Peritoneal Glucose Absorption, and Estimation of Dietary Protein Intake
Adequacy of dialysis was calculated from 24-hour urine and dialysate collection. Weekly Kt/V was determined using standard methods (12). Residual renal function was calculated as the average of 24-hour urinary urea and creatinine, as described elsewhere (13). Small-solute peritoneal membrane transport was calculated from a 4-hour, 3.86% glucose PET according to the two-in-one protocol (14). Glucose drained over 24 hours was measured. Peritoneal glucose absorption was calculated by subtracting the 24-hour drained glucose from the total glucose influx by the dialysate. Dietary protein intake was estimated from protein catabolic rate normalized to actual body weight (nPCR) using the PD Adequest software (Baxter Healthcare Corporation, Deerfield, IL).
Body Composition Assessment
We performed multifrequency whole body bioimpedance assessment using the Body Composition Monitor (Fresenius Medical Care, Bad Homburg, Germany). The bioimpedance method applied was validated by isotope dilution methods (15), by accepted reference body composition methods (16,17), and by extensive clinical assessment of the hydration state (18). Body composition assessments were done with full abdomen during the PET procedure, taking into account the fact that the patient was weighed after draining out peritoneal effluent and thereafter 2 liters of peritoneal solution were instilled.
We measured lean tissue index (LTI), fat tissue index (FTI), relative fat mass (rel.FM), body cell mass (BCM), and relative overhydration (rel.OH). Only measurements simultaneously achieving high quality (>90%) and low error percentage (<35%), showing an inverted U curve in the device monitor, were validated.
Due to biophysic reasons, the bioimpedance spectroscopy does not measure sequestered fluid in the trunk (19), and therefore the presence or absence of PD fluid in the abdomen does not influence body composition parameters related to nutritional status (20). Women with rel.FM >30% and men with rel.FM >25% were considered obese (21).
The values for LTI were compared with an age- and gender-matched reference population and, according to this, patients with LTI <10th percentile were considered to have protein wasting (22). Patients that simultaneously presented with rel.FM<10% and LTI <reference range were characterized as having protein-energy wasting (23).
Statistical Analyses
The data are presented as means±SD or as median with the interquartile range. Pearson and Spearman correlation analyses were used, as appropriate, to assess a linear correlation between insulin resistance indices and several metabolic, inflammatory, and body composition parameters, peritoneal glucose absorption, and small-solute transport status.
Continuous variables were compared between groups using t-test or Mann–Whitney U test, as appropriate. The chi-squared test was used for categorical variables. Comparison between obese and nonobese patients was performed according to body mass index (BMI) and according to rel.FM by bioimpedance analysis. We also performed comparisons concerning fast transport status and insulin resistance state.
Independent correlates of HOMA-IR were assessed by multivariate analysis, considering the significant associated variables recognized by bivariate analysis, and adjusted for peritoneal glucose absorption, small-solute transport, and time on PD. LAR, C-reactive protein, and leptin were natural log-transformed due to the skewed distribution. Values of P<0.05 were considered statistically significant. Statistical analyses were performed using SPSS software application (version 21; SPSS, Chicago, Illinois).
Results
General Features of the Study Population
Table 1 lists features of the study population. Mean dialysate to plasma ratio (D/P)creatinine was 0.8±0.1; 13.7% were fast transporters. Mean fasting glucose and insulin were within normal ranges. Insulin resistance was detected in 18 patients (35.3%) according to HOMA-IR. Using BMI, 20 patients (39%) were considered overweight or obese. According to rel.FM, 25 patients (49%) were obese. The nutritional spectrum included nine obese patients with sarcopenia (LTI <reference range) and two with protein-energy wasting (rel.FM<10% and LTI <reference range).
Table 1.
Characteristics of the study patients
| Variable | Mean±SD or Median (Interquartile Range) |
|---|---|
| Age, years | 50.4±15.9 |
| Time on PD, months | 36.3±33.6 |
| Sex (female), % | 58.8 |
| APD/CAPD | 22/29 |
| HOMA-IR | 2.1±1.4 |
| Fasting plasma insulin, μU/ml | 9.6±5.5 |
| Fasting plasma glucose, mg/dl | 84.9±11.8 |
| HOMA-AD | 42.0±33.6 |
| IGFBP-1, ng/ml | 16.0±14.6 |
| Adiponectin, mcg/ml | 24.3±7.9 |
| Leptin (ng/ml) | 29.2 (64.2) |
| LAR | 1.0 (3.3) |
| Albumin, g/dl | 3.9±0.4 |
| Prealbumin, mg/dl | 34.2±7.6 |
| nPCR, g/kg per day | 1.1±0.3 |
| C-reactive protein (mg/L) | 3.1 (5.7) |
| D/Pcreatinine | 0.8±0.1 |
| Weekly Kt/V urea | 2.2±0.5 |
| RRF, ml/min per 1.73 m2 | 3.1±2.9 |
| BMI, kg/m2 | 24.0±4.1 |
| FTI, kg/m2 | 9.8±5.4 |
| rel.FM, % | 28.8±12.3 |
| LTI, kg/m2 | 13.9±3.7 |
| rel.OH, % | 6.4±9.1 |
| Glucose load, g/day | 146.5±89.2 |
| Glucose absorbed, g/day | 60.1±29.5 |
| Causes of renal failure | Number of patients |
| Chronic glomerulonephritis (%) | 28 (55) |
| Tubulointerstitial nephritis (%) | 5 (9.8) |
| ADPRD (%) | 4 (7.8) |
| CAKUT (%) | 3 (5.9) |
| Chronic pyelonephritis (%) | 3 (5.9) |
| CKD of unknown etiology (%) | 8 (15.7) |
Values are expressed as frequencies for categorical variables and as mean±SD or median (interquartile range) for continuous variables. PD, peritoneal dialysis; APD, automated peritoneal dialysis; CAPD, continuous ambulatory peritoneal dialysis; HOMA-IR, homeostasis model assessment insulin resistance index; HOMA-AD, homeostasis model assessment corrected for adiponectin; IGFBP-1, insulin-like growth factor binding protein 1; LAR, leptin/adiponectin ratio; nPCR, normalized protein catabolic rate; D/P, dialysate to plasma ratio; RRF, residual renal function; BMI, body mass index; FTI, fat tissue index; rel.FM, relative fat mass; LTI, lean tissue index; rel.OH, relative overhydration; ADPRD, autosomal dominant polycystic renal disease; CAKUT, congenital anomalies of the kidney and urinary tract.
The prevalence of obesity (according to rel.FM) was significantly higher in patients on PD for >1 year (61%), than in patients on PD for <1 year (20%) (odds ratio [OR]=6.29; 95% CI, 1.50 to 26.31; P=0.01).
Correlations between Insulin Resistance Indices, Small-Solute Transport, Glucose Absorption, and Body Composition Parameters
HOMA-IR correlated better with FTI than with BMI both in the general population and in the obese group (Table 2). HOMA-IR correlated strongly with HOMA-AD. Notably, significant correlations were found between HOMA-IR and leptin, and LAR and IGFBP-1 in the obese group, but not in the nonobese group (Table 2). Instead of the above positive correlation in the obese group, LAR<1 was documented in the two patients with protein-energy wasting, with high HOMA-IR.
Table 2.
Correlations between HOMA-IR and other insulin resistance indices (HOMA-AD and LAR), small-solute transport (D/Pcreatinine), glucose absorbed, and body composition parameters
| Variable | All Patients N=51 | Obese Patients According to rel.FM n=25 | Nonobese Patients According to rel.FM n=26 | |||
|---|---|---|---|---|---|---|
| r | P Value | r | P Value | r | P Value | |
| Age | −0.001 | 0.10 | −0.26 | 0.22 | 0.07 | 0.75 |
| Time on PD | −0.15 | 0.30 | −0.23 | 0.26 | −0.34 | 0.19 |
| HOMA-AD | 0.73 | <0.001a | 0.77 | <0.001a | 0.77 | <0.001a |
| IGFBP-1 | −0.43 | 0.01a | −0.56 | 0.01a | −0.28 | 0.25 |
| Leptin | 0.48 | <0.001a | 0.65 | <0.001a | 0.15 | 0.47 |
| Adiponectin | −0.05 | 0.77 | 0.04 | 0.87 | −0.03 | 0.90 |
| LAR | 0.55 | <0.001a | 0.74 | <0.001a | −0.09 | 0.72 |
| D/Pcreatinine | 0.06 | 0.68 | 0.16 | 0.50 | 0.24 | 0.26 |
| Glucose absorbed | −0.14 | 0.33 | −0.08 | 0.72 | −0.22 | 0.29 |
| BMI | 0.44 | 0.001a | 0.43 | 0.03a | 0.16 | 0.44 |
| FTI | 0.46 | 0.001a | 0.54 | 0.01a | −0.19 | 0.36 |
| LTI | −0.14 | 0.34 | −0.06 | 0.78 | 0.32 | 0.11 |
PD, peritoneal dialysis; HOMA-AD, homeostasis model assessment corrected for adiponectin; IGFBP-1, insulin-like growth factor binding protein 1; LAR, leptin/adiponectin ratio; D/P, dialysate to plasma ratio; BMI, body mass index; FTI, fat tissue index; rel.FM, relative fat mass; LTI, lean tissue index.
P<0.05.
There were no correlations between insulin resistance indices and renal function, Kt/V urea, peritoneal transport, or glucose absorption (Table 2). A significant correlation of HOMA-IR with lean mass was not found.
Metabolic and Inflammatory Parameters according to BMI and rel.FM
According to BMI, there were no differences in fasting insulin, HOMA-IR, or HOMA-AD between normal weight and obese patients. However, according to rel.FM, HOMA-IR, HOMA-AD, and fasting insulin were significantly higher in obese patients and the difference in LAR was more pronounced (Table 3). Similarly, C-reactive protein did not differ significantly between patients categorized according to BMI, but was significantly higher in obese patients characterized by rel.FM (3.6 [7.4] versus 1.3 [4.4], P=0.03).
Table 3.
Comparison between overweight/obese and normal weight patients according to BMI and comparison between obese and nonobese patients according to rel.FM
| Variable | Normal Weight N=31 18.5≤BMI<25 kg/m2 | Overweight/Obese n=20 BMI≥25 kg/m2 | P Value | Nonobese n=26 rel.FM<25% (Men), <30% (Women) | Obese n=25 rel.FM≥25% (Men), ≥30% (Women) | P Value |
|---|---|---|---|---|---|---|
| Time on PD, months | 30.1±31.4 | 48.1±34.9 | 0.07 | 28.3±30.8 | 46.2±34.7 | 0.06 |
| Age, years | 45.4±114.8 | 59.0±14.4 | 0.002 | 44.9±14.1 | 56.7±15.8 | 0.01 |
| Glucose, mg/dl | 83.2±12.2 | 88.5±11.0 | 0.12 | 86.0±12.2 | 84.8±11.8 | 0.73 |
| Insulin, μU/ml | 8.6±4.6 | 11.6±6.3 | 0.06 | 8.0±3.6 | 11.6±6.5 | 0.02 |
| HOMA-IR | 1.8±1.1 | 2.6±1.6 | 0.06 | 1.7±0.9 | 2.5±1.6 | 0.04 |
| HOMA-AD | 32.9±24.3 | 51.6±40.9 | 0.09 | 27.5±19.2 | 52.4±39.0 | 0.01 |
| LAR | 0.7 (2.0) | 1.9 (8.9) | 0.01 | 0.4 (0.9) | 2.8 (8.0) | <0.001 |
| IGFBP-1, ng/ml | 18.7±16.6 | 12.0±11.3 | 0.15 | 18.1±17.7 | 13.7±11.8 | 0.37 |
| nPCR, g/kg per day | 1.1±0.2 | 1.0±0.3 | 0.05 | 1.2±0.2 | 1.0±0.1 | 0.01 |
| Albumin, g/dl | 3.9±0.4 | 3.9±0.4 | 0.97 | 3.9±0.4 | 4.0±0.4 | 0.87 |
| C-reactive protein, mg/L | 3.2 (4.4) | 2.8 (10.6) | 0.27 | 1.3 (4.4) | 3.6 (7.4) | 0.03 |
| Adiponectin, mcg/ml | 26.2±7.3 | 23.2±7.2 | 0.21 | 26.9±7.0 | 23.2±7.3 | 0.11 |
| Leptin (ng/ml) | 14.7 (42.7) | 49.8 (169.9) | 0.01 | 12.8 (24.7) | 53.4 (162.6) | <0.001 |
| Weekly Kt/V urea | 2.2±0.5 | 2.1±0.5 | 0.39 | 2.2±0.5 | 2.2±0.5 | 0.80 |
| RRF, ml/min per 1.73 m2 | 3.2±2.5 | 2.8±3.3 | 0.57 | 3.8±3.4 | 2.3±1.8 | 0.05 |
| D/Pcreatinine | 0.8±0.1 | 0.8±0.1 | 0.49 | 0.8±0.1 | 0.7±0.1 | 0.02 |
| Glucose load, g/day | 128.3±74.6 | 168.7±100.4 | 0.14 | 134.5±74.8 | 154.6±98.6 | 0.43 |
| Glucose absorbed, g/day | 55.5±26.6 | 63.6±30.6 | 0.33 | 59.9±25.2 | 57.8±31.4 | 0.80 |
| BMI, kg/m2 | 22.1±1.9 | 28.1±3.2 | <0.001 | 22.9±2.5 | 26.0±4.4 | 0.004 |
| Fat mass, kg | 14.2±3.6 | 27.1±10.2 | <0.001 | 12.1±4.7 | 27.0±8.7 | <0.001 |
| FTI, kg/m2 | 7.4±3.4 | 13.9±5.3 | <0.001 | 6.0±2.3 | 14.0±4.3 | <0.001 |
| rel.FM, % | 24.8±10.9 | 36.0±11.1 | 0.001 | 19.2±7.4 | 39.1±6.7 | <0.001 |
| LTI, kg/m2 | 14.2±3.6 | 13.6±4.0 | 0.63 | 16.6±3.2 | 11.4±2.0 | <0.001 |
| BCM, kg | 21.7±8.4 | 21.0±9.9 | 0.79 | 27.3±8.6 | 15.7±4.7 | <0.001 |
| rel.OH, % | 7.3±9.6 | 4.80±8.06 | 0.34 | 5.1±9.2 | 7.5±8.8 | 0.35 |
Values are expressed as mean±SD or median (interquartile range) as appropriate. BMI, body mass index; rel.FM, relative fat mass; PD, peritoneal dialysis; HOMA-IR, homeostasis model assessment insulin resistance index; HOMA-AD, homeostasis model assessment corrected for adiponectin; LAR, leptin/adiponectin ratio; IGFBP-1, insulin-like growth factor binding protein 1; nPCR, normalized protein catabolic rate; RRF, residual renal function; D/P, dialysate to plasma ratio; FTI, fat tissue index; LTI, lean tissue index; BMC, body cell mass; rel.OH, relative overhydration.
Time on PD, residual renal function, Kt/V, peritoneal glucose absorption, and D/Pcreatinine did not differ significantly between groups, according to BMI and rel.FM.
Fast Transport Status Associations with Insulin Resistance, Glucose Absorption, and Body Composition Parameters
Fast transporters were more frequently prescribed automated PD (57.1% versus 40.9%, P=0.62) and icodextrin (71.4% versus 36.4%, P=0.05). There were no significant differences between fast transporters and nonfast transporters concerning fasting glucose (88.0±9.1 versus 85.0±12.0 mg/dl, P=0.47), fasting insulin (10.5±5.1 versus 7.8±5.7 μU/ml, P=0.63), HOMA-IR (2.3±1.3 versus 2.1±1.4, P=0.55), HOMA-AD (34.7±26.9 versus 45.4±36.2, P=0.37), LAR (3.1 [5.8] versus 3.8 [5.5], P=0.37), and IGFBP-1 (24.5±27.0 versus 12.3±8.0, P=0.55). PD regimens also allowed similar daily peritoneal glucose absorption. There were no differences concerning body composition parameters related both with fat mass, lean mass, and relative overhydration between fast transporters and the other small-solute transport categories. The prevalence of both wasting (28.6% versus 15.4%, P=0.59) and obesity (28.6% versus 48.7%, P=0.43) were not significantly different between the two groups.
Insulin Resistance in Prevalent Patients
Patients with >1 year on PD, classified as insulin resistant (HOMA≥2.2), presented significantly higher values of HOMA-AD and LAR (70.7±40.4 versus 23.5±11.4, and 7.3 [12.3] versus 0.7 [1.4] respectively, both P<0.001], and significantly lower values of IGFBP-1 (8.2±7.2 ng/ml versus 21.0±16.3 ng/ml, P=0.002). Body composition parameters related to fat mass were significantly higher in insulin resistant patients, while parameters related to lean mass were similar between the groups (Table 4).
Table 4.
Differences in laboratory parameters, small-solute transport, glucose absorption, and body composition parameters according to insulin resistance in patients on PD for >1 year
| Variable | HOMA-IR<2.2 n=23 | HOMA-IR≥2.2 n=13 | P Value |
|---|---|---|---|
| Age, years | 53.7±16.7 | 51.9±15.8 | 0.63 |
| Glucose, mg/dl | 83.6±10.9 | 91.9±10.0 | 0.03 |
| Insulin, μU/ml | 6.6±2.1 | 15.8±5.9 | <0.001 |
| HOMA-IR | 1.4±0.4 | 3.6±2.3 | <0.001 |
| HOMA-AD | 23.5±11.1 | 70.7±40.4 | <0.001 |
| Leptin/adiponectin ratio | 0.7 (1.4) | 7.3 (12.3) | <0.001 |
| IGFBP-1, ng/ml | 21.0±16.3 | 8.2±7.2 | 0.002 |
| nPCR, g/kg per day | 1.1±0.3 | 0.9±0.2 | 0.17 |
| Albumin, g/dl | 3.9±0.4 | 4.0±0.5 | 0.39 |
| C-reactive protein ( mg/L) | 2.7 (5.6) | 3.5 (5.1) | 0.75 |
| Adiponectin, mcg/ml | 25.9±7.7 | 22.8±6.9 | 0.30 |
| Leptin (ng/ml) | 14.7 (37.2) | 140.7 (228.1) | <0.001 |
| Weekly Kt/V urea | 2.2±0.5 | 2.0±0.5 | 0.11 |
| RRF, ml/min per 1.73 m2 | 2.5±3.3 | 2.4±1.8 | 0.72 |
| D/Pcreatinine | 0.8±0.1 | 0.8±0.1 | 1.00 |
| Glucose load, g/day | 168.2±87.5 | 162.0±95.4 | 0.75 |
| Glucose absorbed, g/day | 62.0±25.9 | 63.4±35.2 | 0.85 |
| BMI, kg/m2 | 24.3±3.0 | 26.6±5.3 | 0.28 |
| Fat mass, kg | 18.0±8.3 | 28.3±12.0 | 0.01 |
| FTI, kg/m2 | 9.1±4.3 | 14.7±5.9 | 0.004 |
| rel.FM, % | 27.2±11.5 | 39.4±10.1 | 0.002 |
| LTM, kg | 40.2±14.6 | 31.5±10.6 | 0.12 |
| Obesity according to rel.FM (%) | 11/23 (47.8) | 11/13 (84.6) | 0.04 |
| Protein wasting according to LTI (%) | 3/23 (13) | 5/13 (38.5%) | 0.11 |
Values are expressed as mean±SD or median (interquartile range) as appropriate.PD, peritoneal dialysis; HOMA-IR, homeostasis model assessment insulin resistance index; HOMA-AD, homeostasis model assessment corrected for adiponectin; IGFBP-1, insulin-like growth factor binding protein 1; nPCR, normalized protein catabolic rate; RRF, residual renal function; D/P, dialysate to plasma ratio; FTI, fat tissue index; rel.FM, relative fat mass; LTM, lean tissue mass; LTI, lean tissue index.
The prevalence of obesity according to rel.FM was significantly higher in the insulin-resistant group than in patients with HOMA<2.2 (84.6% versus 47.8%, P=0.04). The prevalence of protein wasting according to bioimpedance was similar between the two groups (Table 4). Mean daily peritoneal glucose absorption was not significantly different between the two groups (Table 4).
Determinants of HOMA-IR in PD
In a multivariate linear regression model adjusted for peritoneal glucose absorption, small-solute transport, C-reactive protein, and time on PD, only FTI and LAR were independently correlated with HOMA-IR (r=0.82, P<0.001). HOMA-IR was 0.16 units higher per kg/m2 of FTI (95% CI, 0.05 to 0.26, P=0.01) and 0.10 units higher per unit of LAR (95% CI, 0.01 to 0.2, P=0.03).
Discussion
To the best of our knowledge this is the first study that explores the relationships between HOMA-IR and new adipokines-based insulin resistance indices with peritoneal glucose absorption, small-solute transport, and body composition assessed by bioimpedance in PD patients. It also sheds light on the factors that predict insulin resistance in PD patients, showing the role of obesity and adipokines in this population. PD is associated with an increased risk of cardiovascular mortality after the first year (24). Though speculative, it is feared that glucose absorption could explain this outcome by promoting body fat gain and insulin resistance (2,4). Only two studies have specifically addressed the relationship of HOMA-IR with outcomes in PD (9,25). These studies presented contradictory results and did not address the role of glucose absorption and fast transport status on insulin resistance, a major limitation. In our study, peritoneal glucose absorption was not correlated with any insulin resistance indices, nor with fat mass, and did not differ in insulin-resistant patients versus the lower HOMA-IR group. Our results also reproduce our previous observation that BMI underestimates obesity prevalence in PD (6), and that obesity increases in patients on PD for >1 year (61%). This supports other studies reporting increased fat mass with time on PD (8,26–28). However, none of the body fat mass composition parameters showed correlation with glucose absorption or fast transport status. In line with our finding, others documented a lack of association between peritoneal glucose exposure and body fat gain in patients on PD (7,29).
We also demonstrate that significant differences in HOMA-IR, HOMA-AD, and LAR were only evident when patients were compared according to rel.FM and not according BMI. Again, these results underline the impact of obesity on insulin resistance development in PD patients, as HOMA-IR, HOMA-AD, and LAR were significantly higher in obese patients compared to the results in nonobese patients, when categorized according to rel.FM.
A clinically relevant conclusion is that the relationship between insulin resistance and the adipokines ratio (LAR) is regulated by body composition parameters. We have previously reported that rel.FM and LTI predicted LAR independently of peritoneal glucose absorption (6). The results of the present study corroborate this, underlining that insulin resistance is aggravated in both extremes of the nutritional spectrum: obese patients exhibited higher LAR values, strongly positively correlated with HOMA-IR, but the two patients with protein-energy wasting also evidenced insulin resistance while LAR was <1. We measured IGFBP-1, which has recently been validated as a surrogate marker of visceral fat, and is strongly inversely related to liver fat content measured by the gold standard method, proton magnetic resonance spectroscopy (30). We report a significant negative correlation between HOMA-IR and IGFBP-1 in our population, especially in obese patients (r=–0.56, P=0.01), in line with IGFBP-1 being a marker of hepatic insulin sensitivity.
Concerning fast transport status, it is feared that higher glucose absorption could induce metabolic alterations and cardiovascular events (2). In a previous study of PD patients followed up on for 47 months, no difference was found concerning small-solute transport rate between patients with and without cardiovascular events (6). In the present study, we also did not find any significant difference concerning LAR, IGFPB-1, glucose absorption, or body composition parameters between fast transporters and other categories of small-solute transport. Only two studies have assessed the correlations between peritoneal transport and insulin resistance measured by HOMA-IR (31,32), concluding that D/Pcreatinine was not significantly associated with HOMA-IR. Our results not only support such evidence, but expand on the knowledge regarding fast transport status systemic consequences, highlighting the fact that fast transporters under updated therapy regimens have no higher likelihood of obesity or overhydration compared with nonfast transporters. The use of new solutions, icodextrin, and short dwell times could contribute to minimizing glucose exposure and systemic effects.
Also relevant was to document that patients categorized as insulin resistant presented significantly higher adipokine-based insulin resistance indices, lower IGFBP-1, and significantly higher values of body composition parameters related with fat mass, in spite of similar BMI scores.
Leptin and adiponectin play important and opposite roles in the regulation of cardiovascular and metabolic homeostasis. We recently reported that LAR can be used as a new atherogenic index, as it is associated with cardiovascular events in PD patients without protein wasting (6). According to our present investigation this association between LAR and cardiovascular events might be explained through development of insulin resistance. In fact, given the direct actions of adiponectin on insulin sensitivity both in skeletal muscle and liver (33,34), a higher LAR can been seen not only as a biomarker of insulin resistance, but also as a predictor of insulin resistance. Our study also highlights the fact that obesity is associated with insulin resistance in PD patients, independently of peritoneal glucose absorption and fast transport status.
Our study is limited by its cross-sectional design and sample size, and for that reason we aim for a longitudinal extension of this investigation. Validation of insulin resistance indices against the euglycemic hyperinsulinemic clamp in the PD field would be valuable, since it is regarded as the reference method for insulin sensitivity assessment. However, this method is expensive and not suitable for use in clinical practice. HOMA-IR, however, is easy to perform and validated as a valuable alternative of insulin sensitivity both in non-CKD patients (11,35), in CKD patients stages III and IV (36,37), and in HD patients (10).
There are also strengths to this study. The few studies that have calculated HOMA-IR in PD patients lack any standardization of PD prescription concerning the last dwell period, before collecting blood to analyze insulin and glucose (9,25). We standardized these procedures not only in a fasting state, but also after the overnight dwell with 1.36% glucose solution, before the PET. Such a protocol minimizes any potential effect of peritoneal glucose absorption, while allowing for dialysis maintenance, as a previous study by Heaton et al. showed that there are no differences in fasting plasma glucose and insulin between undialyzed PD patients (with no PD solution in the abdomen) and patients on 1.36% glucose solution, after the first 60 minutes of a 1.36% glucose dwell (38). Assuming this, other investigators focused on metabolic issues, performing measurements after the 1.36% glucose solution overnight dwell (39,40).
Insulin assay can vary considerably, especially if antibodies that crossreact with insulin or split-proinsulin products are used (41): we used an insulin-specific assay, thereby minimizing the interference exerted on the HOMA-IR score. The systemic impact of peritoneal glucose absorption is scarcely addressed, and the few studies about insulin resistance in PD do not take into account the impact of protein-energy wasting and obesity in its development. In the present investigation, we have not only measured peritoneal glucose absorption, but also assessed body composition by bioimpedance method. This may suffer from methodologic errors, but is validated as a clinical tool to evaluate body composition in ESRD patients (19,20,42,43).
To summarize, our results evidence that obesity and adipocytokines profile are associated with insulin resistance in PD, independently of glucose absorption and small-solute transport status. Patients with insulin resistance, namely with higher LAR, are more prone to suffer from cardiovascular events (6). In order to possibly break such a chain of events, efforts should be put into obesity prevention, as well as correct diagnosis and treatment. One of the most feared systemic consequences of glucose absorption in PD is obesity and/or development of insulin resistance. However, fast transport status under contemporary PD regimens was not associated with obesity or insulin resistance. Current PD treatments minimize glucose exposition, and the lack of correlations between peritoneal glucose absorption and insulin resistance indices and body composition parameters puts forward evidence that other factors may contribute to a metabolic syndrome development in PD patients, perhaps in a more powerful way. Beyond genetics, future efforts should be made towards determining resting energy expenditure (44) and the role of energy intake and physical exercise.
Disclosures
None.
Acknowledgments
This study was partially performed with the help of investigational grants from Sociedade Portuguesa de Nefrologia. This work was also supported by the Unit for Multidisciplinary Research in Biomedicine, Abel Salazar Biomedical Sciences Institute, University of Porto. The Unit for Multidisciplinary Research in Biomedicine is funded by grants from Foundation for Science and Technology (Fcomp-01-0124-FEDER-015896).
The results presented in this paper have not been published previously in whole or part.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
References
- 1.Sechi LA, Catena C, Zingaro L, Melis A, De Marchi S: Abnormalities of glucose metabolism in patients with early renal failure. Diabetes 51: 1226–1232, 2002 [DOI] [PubMed] [Google Scholar]
- 2.Fortes PC, de Moraes TP, Mendes JG, Stinghen AE, Ribeiro SC, Pecoits-Filho R: Insulin resistance and glucose homeostasis in peritoneal dialysis. Perit Dial Int 29[Suppl 2]: S145–S148, 2009 [PubMed] [Google Scholar]
- 3.Kobayashi S, Maejima S, Ikeda T, Nagase M: Impact of dialysis therapy on insulin resistance in end-stage renal disease: comparison of haemodialysis and continuous ambulatory peritoneal dialysis. Nephrol Dial Transplant 15: 65–70, 2000 [DOI] [PubMed] [Google Scholar]
- 4.Kim YL, Cho JH, Choi JY, Kim CD, Park SH: Systemic and local impact of glucose and glucose degradation products in peritoneal dialysis solution. J Ren Nutr 23: 218–222, 2013 [DOI] [PubMed] [Google Scholar]
- 5.Bernardo AP, Fonseca I, Rodrigues A, Carvalho MJ, Cabrita A: Overweight rather than malnutrition is widely prevalent in peritoneal dialysis patients. Adv Perit Dial 25: 119–124, 2009 [PubMed] [Google Scholar]
- 6.Bernardo AP, Fonseca I, Oliveira JC, Santos O, Carvalho MJ, Cabrita A, Rodrigues A: Adipokines in peritoneal dialysis: relevant clinical impact according to body composition. Ther Apher Dial 19: 144–153, 2015 [DOI] [PubMed] [Google Scholar]
- 7.Vasselai P, Kamimura MA, Bazanelli AP, Pupim LB, Avesani CM, da Mota Ribeiro FS, Manfredi SR, Draibe SA, Cuppari L: Factors associated with body-fat changes in prevalent peritoneal dialysis patients. J Ren Nutr 18: 363–369, 2008 [DOI] [PubMed] [Google Scholar]
- 8.Fernström A, Hylander B, Moritz A, Jacobsson H, Rössner S: Increase of intra-abdominal fat in patients treated with continuous ambulatory peritoneal dialysis. Perit Dial Int 18: 166–171, 1998 [PubMed] [Google Scholar]
- 9.Li Y, Zhang L, Gu Y, Hao C, Zhu T: Insulin resistance as a predictor of cardiovascular disease in patients on peritoneal dialysis. Perit Dial Int 33: 411–418, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hung AM, Sundell MB, Egbert P, Siew ED, Shintani A, Ellis CD, Bian A, Ikizler TA: A comparison of novel and commonly-used indices of insulin sensitivity in African American chronic hemodialysis patients. Clin J Am Soc Nephrol 6: 767–774, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bonora E, Targher G, Alberiche M, Bonadonna R, Saggiani F, Zenere MB, Monauni T, Muggeo M: Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 23: 57–63, 2000 [DOI] [PubMed] [Google Scholar]
- 12.Nolph KD, Moore HL, Twardowski ZJ, Khanna R, Prowant B, Meyer M, Ponferrada L: Cross-sectional assessment of weekly urea and creatinine clearances in patients on continuous ambulatory peritoneal dialysis. ASAIO J 38: M139–M142, 1992 [DOI] [PubMed] [Google Scholar]
- 13.van Olden RW, Krediet RT, Struijk DG, Arisz L: Measurement of residual renal function in patients treated with continuous ambulatory peritoneal dialysis. J Am Soc Nephrol 7: 745–750, 1996 [DOI] [PubMed] [Google Scholar]
- 14.Bernardo AP, Bajo MA, Santos O, del Peso G, Carvalho MJ, Cabrita A, Selgas R, Rodrigues A: Two-in-one protocol: simultaneous small-pore and ultrasmall-pore peritoneal transport quantification. Perit Dial Int 32: 537–544, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Moissl UM, Wabel P, Chamney PW, Bosaeus I, Levin NW, Bosy-Westphal A, Korth O, Müller MJ, Ellegård L, Malmros V, Kaitwatcharachai C, Kuhlmann MK, Zhu F, Fuller NJ: Body fluid volume determination via body composition spectroscopy in health and disease. Physiol Meas 27: 921–933, 2006 [DOI] [PubMed] [Google Scholar]
- 16.Wabel P, Chamney P, Moissl U: Reproducibility of bioimpedance spectroscopy for the assessment of body composition and dry weight. J Am Soc Nephrol 18: A255, 2007 [Google Scholar]
- 17.Moissl U, Wabel P, Chamney PW: Validation of bioimpedance spectroscopy method for the assessment of fat free mass. NDT Plus 1(Suppl2):ii215.XLV Congress of the ERA-EDTA, 2008 [Google Scholar]
- 18.Chamney PW, Kramer M, Rode C, Kleinekofort W, Wizemann V: A new technique for establishing dry weight in hemodialysis patients via whole body bioimpedance. Kidney Int 61: 2250–2258, 2002 [DOI] [PubMed] [Google Scholar]
- 19.Cooper BA, Aslani A, Ryan M, Zhu FY, Ibels LS, Allen BJ, Pollock CA: Comparing different methods of assessing body composition in end-stage renal failure. Kidney Int 58: 408–416, 2000 [DOI] [PubMed] [Google Scholar]
- 20.Arroyo D, Panizo N, Abad S, Vega A, Rincón A, de José AP, López-Gómez JM. Intraperitoneal fluid overestimates hydration status assessment by bioimpedance spectroscopy. Perit Dial Int 35: 85–89, 2015 [DOI] [PMC free article] [PubMed]
- 21.American Society of Bariatric Physicians: Overweight and obesity evaluation and management. Available at: http://www.asbp.org/siterun_data/about_asbp/position_statements/doc7270523281295654373.html. Accessed February, 14, 2012
- 22.Wieskotten S, Heinke S, Wabel P, Moissl U, Becker J, Pirlich M, Keymling M, Isermann R: Bioimpedance-based identification of malnutrition using fuzzy logic. Physiol Meas 29: 639–654, 2008 [DOI] [PubMed] [Google Scholar]
- 23.Fouque D, Kalantar-Zadeh K, Kopple J, Cano N, Chauveau P, Cuppari L, Franch H, Guarnieri G, Ikizler TA, Kaysen G, Lindholm B, Massy Z, Mitch W, Pineda E, Stenvinkel P, Treviño-Becerra A, Wanner C: A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic kidney disease. Kidney Int 73: 391–398, 2008 [DOI] [PubMed] [Google Scholar]
- 24.Johnson DW, Dent H, Hawley CM, McDonald SP, Rosman JB, Brown FG, Bannister K, Wiggins KJ: Association of dialysis modality and cardiovascular mortality in incident dialysis patients. Clin J Am Soc Nephrol 4: 1620–1628, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sanchez-Villanueva R, Estrada P, del Peso G, Grande C, Diez JJ, Iglesias P, Gonzalez E, Aguilar-Rodriguez A, Selgas R, Bajo MA, Grupo de Estudios Peritoneales de Madrid de REDINREN (Red Renal de Investigación de la RETICS 06/0016, del Instituto de Salud Carlos III) del IRSIN (Instituto Reina Sofía de Investigación Nefrológica) : Repeated analysis of estimated insulin resistance using the HOMAIR index in nondiabetic patients on peritoneal dialysis and its relationship with cardiovascular disease and mortality. Nefrologia 33: 85–92, 2013 [DOI] [PubMed] [Google Scholar]
- 26.Stenvinkel P, Lindholm B, Lönnqvist F, Katzarski K, Heimbürger O: Increases in serum leptin levels during peritoneal dialysis are associated with inflammation and a decrease in lean body mass. J Am Soc Nephrol 11: 1303–1309, 2000 [DOI] [PubMed] [Google Scholar]
- 27.Søreide R, Dracup B, Svarstad E, Iversen BM: Increased total body fat during PD treatment. Adv Perit Dial 8: 173–176, 1992 [PubMed] [Google Scholar]
- 28.Cho KH, Do JY, Park JW, Yoon KW: Effect of icodextrin dialysis solution on body weight and fat accumulation over time in CAPD patients. Nephrol Dial Transplant 25: 593–599, 2010 [DOI] [PubMed] [Google Scholar]
- 29.Fan S, Davenport A: Does increased glucose exposure lead to increased body fat and reduced lean body mass in anuric peritoneal dialysis patients? Eur J Clin Nutr 68: 1253–1254, 2014 [DOI] [PubMed] [Google Scholar]
- 30.Kotronen A, Lewitt M, Hall K, Brismar K, Yki-Järvinen H: Insulin-like growth factor binding protein 1 as a novel specific marker of hepatic insulin sensitivity. J Clin Endocrinol Metab 93: 4867–4872, 2008 [DOI] [PubMed] [Google Scholar]
- 31.Chen HY, Kao TW, Huang JW, Chu TS, Wu KD: Correlation of metabolic syndrome with residual renal function, solute transport rate and peritoneal solute clearance in chronic peritoneal dialysis patients. Blood Purif 26: 138–144, 2008 [DOI] [PubMed] [Google Scholar]
- 32.da Silva DR, Figueiredo AE, Antonello IC, Poli de Figueiredo CE, d’Avila DO: Solutes transport characteristics in peritoneal dialysis: variations in glucose and insulin serum levels. Ren Fail 30: 175–179, 2008 [DOI] [PubMed] [Google Scholar]
- 33.Rasouli N, Kern PA: Adipocytokines and the metabolic complications of obesity. J Clin Endocrinol Metab 93[Suppl 1]: S64–S73, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kershaw EE, Flier JS: Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 89: 2548–2556, 2004 [DOI] [PubMed] [Google Scholar]
- 35.Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC: Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28: 412–419, 1985 [DOI] [PubMed] [Google Scholar]
- 36.Shoji T, Emoto M, Nishizawa Y: HOMA index to assess insulin resistance in renal failure patients. Nephron 89: 348–349, 2001 [DOI] [PubMed] [Google Scholar]
- 37.Jia T, Huang X, Qureshi AR, Xu H, Ärnlöv J, Lindholm B, Cederholm T, Stenvinkel P, Risérus U, Carrero JJ: Validation of insulin sensitivity surrogate indices and prediction of clinical outcomes in individuals with and without impaired renal function. Kidney Int 86: 383–391, 2014 [DOI] [PubMed] [Google Scholar]
- 38.Heaton A, Johnston DG, Burrin JM, Orskov H, Ward MK, Alberti KGMM, Kerr DNS: Carbohydrate and lipid metabolism during continuous ambulatory peritoneal dialysis (CAPD): the effect of a single dialysis cycle. Clin Sci (Lond) 65: 539–545, 1983 [DOI] [PubMed] [Google Scholar]
- 39.Li PK, Kwan BC, Szeto CC, Ko GT: Metabolic syndrome in peritoneal dialysis patients. NDT Plus 1: 206–214, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Szeto CC, Chow KM, Kwan BC, Chung KY, Leung CB, Li PK: New-onset hyperglycemia in nondiabetic chinese patients started on peritoneal dialysis. Am J Kidney Dis 49: 524–532, 2007 [DOI] [PubMed] [Google Scholar]
- 41.Robbins DC, Andersen L, Bowsher R, Chance R, Dinesen B, Frank B, Gingerich R, Goldstein D, Widemeyer HM, Haffner S, Hales CN, Jarrett L, Polonsky K, Porte D, Skyler J, Webb G,Kallagher K: Report of the American Diabetes Association’s task force on standardization of the insulin assay. Diabetes 45: 242–256, 1996 [DOI] [PubMed] [Google Scholar]
- 42.Parmentier SP, Schirutschke H, Schmitt B, Schewe J, Herbrig K, Pistrosch F, Passauer J: Influence of peritoneal dialysis solution on measurements of fluid status by bioimpedance spectroscopy. Int Urol Nephrol 45: 229–232, 2013 [DOI] [PubMed] [Google Scholar]
- 43.Lindley E, Devine Y, Hall L, Cullen M, Cuthbert S, Woodrow G, Lopot F: A ward-based procedure for assessment of fluid status in peritoneal dialysis patients using bioimpedance spectroscopy. Perit Dial Int 25[Suppl 3]: S46–S48, 2005 [PubMed] [Google Scholar]
- 44.Wang X, Axelsson J, Nordfors L, Qureshi AR, Avesani C, Barany P, Schalling M, Heimbürger O, Lindholm B, Stenvinkel P: Changes in fat mass after initiation of maintenance dialysis is influenced by the uncoupling protein 2 exon 8 insertion/deletion polymorphism. Nephrol Dial Transplant 22: 196–202, 2007 [DOI] [PubMed] [Google Scholar]
