The role of abdominal adiposity and insulin resistance in dyslipidemia of chronic renal failure

The Role of Abdominal Adiposity and Insulin Resistance in Dyslipidemia of Chronic Renal Failure P. Lee, MBBS, FRACP, D. O'Neal, MBBS, FRACP, B. Murphy, PhD, FRACP, and J. Best, MD, FRACP, FRCPath • The atherogenic profile of high triglyceride, reduced high-density lipoprotein (HDL) cholesterol, and small lowdensity !ipoprotein particle size found in patients on chronic hemodialysis is known to be associated with insulin resistance and abdominal obesity in the general population. To assess the influence of insulin resistance and abdominal adiposity on the lipid profile in subjects on hemodialysis, intravenous glucose tolerance test and dualenergy x-ray absorptiometry were performed in 26 nondiabetic subjects on hemodialysis and compared with 22 nondiabetic control subjects matched for age, sex, and body mass index. Subjects on hemodialysis were found to have higher triglyceride (133 mg/dL [95% confidence interval, 115 to 159 mg/dL] v 97 mg/dL [95% confidence interval, 80 to 124 mg/dL]; P < 0.05), lower HDL cholesterol (36 -- 3 mg/dL v 51 ± 4 mg/dL [mean ± SEM]; P < 0.01), enhanced insulin response to glucose (2.72 ± 0.28 mUL-~min per mg dL -1 v 1.67 ± 0.22 m U L - l m i n per mg dL-~; P < 0.01), and reduced sensitivity to the action of insulin (2.24 min -~ per mUL -~ min [95% confidence interval, 1.86 to 2.75 min -~ per mUL -1 min] v 4.17 min -~ per mUL -~ min [95% confidence interval, 2.95 to 5.9 min ~ per mUL -~ min]; P < 0.01) than the control subjects. Abdominal adiposity measured by dual-energy x-ray absorptiometry (2,004 ± 210 g v 2,163 ± 198 g [mean ± SEM]; P = NS) and percentage of body fat distributed to the abdomen (10.5% ± 0.3% v 9.7% ± 0.5% [mean ± SEM]; P = NS) did not differ between the t w o groups. Subjects on hemodialysis were insulin resistant, but unlike control subjects, their lipid profile was not predicted by their insulin sensitivity. Abdominal adiposity was associated with a deteriorating lipid profile and insulin resistance in subjects on hemodialysis, as it was in control subjects. The presence of renal failure resulted in additional insulin resistance and a higher triglyceride level in the leaner subjects on hemodialysis compared with control subjects with similar levels of abdominal fat. In the more obese subjects, insulin sensitivity and triglyceride level did not differ between the t w o groups of subjects, although HDL cholesterol level remained low in subjects on hemodialysis. In conclusion, insulin resistance in subjects on hemodialysis did not directly account for their abnormal lipid profile. The negative impact of abdominal adiposity on the metabolic profile was preserved in subjects on hemodialysis, but the presence of renal failure itself resulted in insulin resistance in the leaner subjects and dyslipidemia in all subjects on hemodialysis compared with control subjects of comparable abdominal adiposity.

© 1997 by the National Kidney Foundation, Inc. INDEX WORDS: Dysiipidemia; insulin resistance; body fat distribution.

D

YSLIPIDEMIA accompanies chronic renal disease v3 and represents an important cardiovascular risk factor4 for patients on chronic dialysis. In subjects on chronic hemodialysis, the lipid profile is characterized by mild hypertriglyceridemia5-7 accompanied by reduced high-density lipoprotein (HDL) cholesterol. 8'9 Although low-density lipoprotein (LDL) cholesterol is not raised in subjects on hemodialysis, a'l°

From the Departments of Medicine and Nephrology, St Vincent's Hospital, Fitzroy, Australia. Received May 9, 1996; accepted in revised form August 13, 1996. Dr Lee was the recipient of a National Health and Medical Research Council of Australia Postgraduate Medical Scholarship. Dr 0 Weal was the recipient of a Merck Sharp and Dohme Lipid Research Fellowship. Address reprint requests to J. Best, MD, FRACP, FRCPath, Department of Medicine, University of Melbourne, St Vincent's Hospital, Fitzroy, Australia 3065. © 1997 by the National Kidney Foundation, Inc. 0272-6386/97/2901-000653.00/0

54

LDL particle size is reduced in the presence of hypertriglyceridemia, u which could potentially increase the atherogenicity of LDL particles 12 and contribute to the risk of cardiovascular disease in these patients. The profile of hypertriglyceridemia, reduced HDL cholesterol, and small LDL particle size has been linked with insulin resistance, ~3 which also is a feature of chronic renal failure. I4'15Since insulin has an important role in lipid metabolism, ~6 resistance to its action is thought to be an important explanation for the dyslipidemia I7'~8 and cardiovascular risk 4 displayed by patients with chronic renal failure. Disturbed glucose-insulin homeostasis and dyslipidemia often coexist with obesity 19 and contribute to the increased risk of cardiovascular disease in obese subjects. Abdominal adiposity in particular is the component of body fat most closely correlated with the risk of cardiovascular disease, 2°-23 dyslipidemia, 24'25 and glucose intolerance. 26 Generalized obesity is not a feature of

American Journal of Kidney Diseases, Vol 29, No 1 (January), 1997: pp 54-65

DYSLIPIDEMIA IN CHRONIC RENAL FAILURE

55

Table 1. Clinical Features of Subjects on Hemodialysis and Control Subjects

corticosteroids, anabolic steroids, testosterone, estrogen, or progestogen therapy. Approval for the study was obtained from the St Vincent' s Hospital Human Research Ethics Committee, and all subjects gave their informed consent. Subjects on hemodialysis were dialyzed for 12 to 13.5 hours per week using cellulose acetate dialyzers against a dextrose-free dialysate. They were prescribed sufficient calories to maintain body weight, with fat intake constituting no more than 30% of total calorie intake and containing equal proportions of saturated, unsaturated, and polyunsaturated fat. Their prescribed protein intake was 1.0 to 1.2 g/kg body weight/d.

No. Age (yr), _+ SEM BMI (kg/m2), _+ SEM Male (%) SBP (mm Hg), _+_ SEM DBP (mm Hg), _+ SEM No. of patients on Betablockers ACE inhibitors Calcium channel antagonists Thiazide Dialysis (mo), _+ SEM

Hemodialysis Subjects

Control Subjects

Probability Value

26 57.5 _+ 2.2

22 53.3 +_ 2.9

NS

26.1 +_ 4.2 88

26.0 + 4.4 68

NS NS

146 _+ 4

128 _+ 4

<0.01

80_+2

82 + 2

NS

Assessment of Insulin Sensitivity

9

3

NS

4 7 0

1 4 2

NS NS NS

29 _+ 8

--

--

Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; ACE, angiotensin-converting enzyme.

r e n a l f a i l u r e , b u t it is n o t k n o w n w h e t h e r a b d o m i nal adiposity plays a role in determining the dyslipidemia and insulin resistance of renal disease o r w h e t h e r t h e d i s t r i b u t i o n o f b o d y f a t to t h e a b d o m e n is a l t e r e d i n r e n a l f a i l u r e . T h e a i m o f t h e p r e s e n t s t u d y w a s to e x a m i n e the influence of insulin resistance on the lipid p r o f i l e o f s u b j e c t s o n h e m o d i a l y s i s a n d to d e t e r m i n e t h e r o l e o f b o d y f a t d i s t r i b u t i o n in t h i s r e l a tionship. MATERIALS AND METHODS

Subjects Twenty-six medically stable subjects on chronic hemodialysis were studied and compared with 22 control subjects matched for age, sex, and body mass index (BMI). Thirteen of the subjects on hemodialysis had been included in a recent study of LDL particle size distribution in renal failure, t~ All subjects on hemodialysis had a stable postdialysis weight and all subjects were nondiabetic. Subjects on hemodialysis were recruited from the Department of Nephrology at St Vincent's Hospital, and control subjects comprised volunteers from hospital staff and the Rotary International organization. Clinical profiles of the two groups are summarized in Table 1. None of the subjects was taking lipid-modifying medications,

After an overnight 12-hour fast, an intravenous glucose tolerance test (IVGTT) was performedY Through an intravenous cannula inserted in one forearm, subjects received a glucose bolus, 0.3 g/kg body weight, over 1 minute. Blood was collected into EDTA-containing tubes at -10, 0, 2, 4, 6, 8, 10, 12, 14, 19, 25, 30, and 40 minutes from an intravenous cannula in the opposite forearm and immediately put on ice. Plasma was separated by centrifugation at 1,500g at 4°C within 2 hours of collection, and aliquots were frozen immediately at -80°C until the time of insulin and nonesterifled fatty acid (NEFA) assays. Figure 1 shows the mean plasma glucose and insulin profiles of subjects on hemodialysis and control subjects following injection of the glucose bolus at 0 minutes. First phase insulin response was calculated as insulin area above basal at 0 to 10 minutes divided by glucose peak above basal at 0 to 10 minutes (mUL-lmin per mg dL-1). Insulin sensitivity was calculated as Kg divided by insulin area above basal at 0 to 40 minutes (min -1 per mUL -~ min), where Kg represents the slope of log glucose concentration between 10 to 40 minutes after glucose bolus. Plasma insulin was measured by radioimmunoassay with dextran charcoal separation of bound and free fractions. 28 Anti-insulin antibody, obtained from Linco Research lnc (St Louis, MO), has no cross-reactivity with either proinsulin or the primary split form, des 31,32 human proinsulin. The interassay coefficient of variation (CV) for plasma insulin was 7,7% at 6.0 mU/L and 6.8% at 26.2 mU/L. Plasma glucose was determined within 12 hours of the IVGTT by a glucose oxidase method with an interassay CV of 2%. Nonesterified free fatty acid was determined in plasma obtained from blood collected at - 1 0 minutes. An enzymatic colorimetric assay (Wako NEFA C, Osaka, Japan) was used as previously described. 29 The interassay CV was 2%. StatView (Abacus Concepts Inc; Berkeley, CA), a standard computerized statistical package, was used to calculate the first phase insulin response, the insulin sensitivity, and Kg, using the trapezoid rule to calculate the area under the insulin curve,

Body Fat Distribution and Quantitation Body fat distribution and quantitation were determined by dual-energy X-ray absorptiometry. A total body scan was performed using a whole body scanner (DPX; Lunar Radiation Corp, Madison, WI), with scan time between 10 and 20

56

LEE ET AL

minutes and the radiation dose between 0.02 and 0.06 torero. Scans were performed with the arms positioned away from the sides of the body to enable an area bound by the first and the fourth lumbar intervertebral discs and the lateral soft tissue margins of the body to be marked on the scan and analyzed using the manual analysis function of the Lunar software. This area was designated the abdomen. In addition, using default Lunar software, the body was separated into an upper and a lower body region by oblique lines passing through the femoral neck and intersecting at the pubic symphysis. The lower body region includes the buttocks and lower limbs, while the upper body region includes all tissues above the oblique lines. Total body tissue, total body fat, fat within the manually defined abdominal region, and fat within the lower body region defined by default software were measured and reported in grams. The right subscapular, paraumbilical, and suprailiac skinfold thicknesses were measured with the same Harpenden skinfold calipers to the nearest 0.1 mm 4 seconds after appli-

300 cv

250 2°0

c~

50 I

0

-10

0

I

I

I

I

10

20

30

40

20

30

40

Minutes

100 80

60 40

-I0

0

10

Minutes

Fig 1. The mean glucose and insulin responses of subjects on hemodialysis (solid diamonds) and control subjects (open circles) following injection of the 0.3 g/kg glucose bolus at O minutes.

cation. The CV was 8.9% for the subscapular, 10.9% for the paraumbilical, and 10.8% for the suprailiac skinfold. Waist to hip ratio was determined with a tape measure to the nearest centimeter at the smallest standing horizontal circumference between the ribs and the iliac crest in expiration, and at the largest standing horizontal circumference of the buttocks, respectively. The CV was 4%. All anthropometric measurements were done by the same investigator (P.L.).

Routine Biochemistry, Parathyroid Hormone, Lipids, and Lipoproteins Prior to commencement of the IVGTT, blood was collected into heparinized tubes for measurement of routine biochemistry by automated laboratory methods. Hemoglobin AIc (HbAIc) was measured with high-performance liquid chromatography 3° with an interassay CV of 5%. In addition, blood also was collected into EDTA-containing tubes prior to the injection of glucose for lipid and lipoprotein analyses. Plasma was separated by centrifugation at 1,500g for 15 minutes at 4°C. Plasma lipids, lipoproteins, and apolipoproteins were measured as previously described. H Lipoprotein AI (LpAI) was measured by means of differential immunoelectrophoresis using a commercial kit (Sebia, Moulineanx, France). Apolipoprotein AI associated with apolipoprotein AII was calculated as the difference between total apolipoprotein AI and the apolipoprotein AI in LpAI. The intergel CV for LpAI was 5.2%. Preparation of LDL was based on an adaptation of the method of Chung et al.3~ Following a 12-hour fast, 10 mL of whole blood was collected from subjects into an EDTAcontaining tube and was centrifuged at 1,500g for 15 minutes to separate the plasma. The density of 4 mL of the plasma was adjusted to 1.21 g/mL by the addition of 1.31 g of KBr. This plasma was then pipetted into an ultracentrifuge tube and overlayered with a 1.006 g/mL solution. The tubes were sealed and placed in a Beckman VTi65.1 rotor (Beckman, Palo Alto, CA) and spun at 65,000 rpm for 90 minutes at 7°C with acceleration and deceleration parameters set at 6. Following ultracentrifugation, the LDL formed a visible band in the middle of the centrifuge tube, with HDL remaining at the bottom of the tube. The ultracentrifuge tube was punctured and the LDL aspirated. The LDL diameter was determined as previously described by Krauss and Burke 32 using commercially available 3% to 13% nondenaturing polyacrylamide gradient gels (Gradipore Ltd, Sydney, Australia). H All gels used in this study were from the same production run. Markers used were 29 nm latex beads (Duke Scientific, Palo Alto, CA) and Pharmacia high molecular weight standards (Pharmacia, Piscataway, NJ). The gels were scanned by a Tracktel Video Densitometer Scanner (Vision System Ltd, Adelaide, Australia) to provide a quantitative measurement of the size of the peak and its distance from the origin. Particle diameter was calculated from a plot of the log of the known diameters of the standards (latex beads 29 nm, thyroglobulin 17 nm, and ferritin 1 2.2 nm) on the y axis against their positions from the origin of the gel on the x axis. A computerized statistical package was used to derive a simple regression equation that allowed samples to be sized from the position each had travelled from

57

DYSLIPIDEMIA IN CHRONIC RENAL FAILURE

Table 2. Lipids, Lipoproteins, Apolipoproteins, Parathyroid Hormone Levels in Subjects on Hemodialysis and Control Subjects

Cholesterol (mg/dL) Tdglyceride (mg/dL) HDL cholesterol (mg/dL) LpAI (g/L) LpAI:AII (g/L) Very LDL cholesterol (mg/dL) LDL cholesterol (mg/dL) LDL diameter (nm) ApoAI (mg/dL) ApoAII (g/L) ApoB (mg/dL) Apo(a) (U/L) Albumin (g/dL) PTH (pg/mL)

Hemodialysis Subjects

Control Subjects

Probability Value

182 _+ 8 133 (95% CI, 115-159) 36 _+ 3 0.42 _+ 0.04 0.67 +_ 0.05 29 _+ 2 116 _+ 7 26,1 + 0.1 112 _+ 8 0.37 + 0.04 108 _+ 7 135 (95% CI, 81-224) 36 _+ 1 105 (95% CI, 62-178)

209 2 8 97 (95%CI, 80-124) 51 + 4 0,58 + 0.05 0.83 +_ 0.05 22 _+ 3 135 _+ 7 26.3 _+ 0.2 143 + 8 0.38 + 0.02 108 _+ 8 129 (95% CI, 107-155) 40 _+ 1 --

<0.05 <0.05 <0.01 <0.05 <0.05 <0.05 <0.05 NS <0.01 NS NS NS <0.001

NOTE. Data are given as mean values +__SEM unless otherwise indicated. Abbreviations: Apo(A), apoprotein (a); ApoAI, apoprotein AI; ApoAII, apoprotein All; ApoB, apoprotein 13; CI, confidence interval; LpAI, lipoprotein AI; LpAI:All, lipoprotein AI:All; PTH, parathyroid hormone.

the origin of the gel. Low-density lipoprotein was obtained from one of the authors (D.C.) and frozen in aliquots at --80°C. Prior to each mn an aliquot was thawed and run on the gel as a quality control. Each aliquot was used once only. ]'he size of this quality control was 26.1 nm and the intergel CV was 0.8%. All subjects on hemodialysis had parathyroid hormone (PTH) measured every 3 months, and their medical records were examined to obtain the PTH levels measured within 3 months of the IVGTr. An immunoradiometricassay (Incstar N-tact PTH SP Kit; Incstar Corp, Stillwater, MN) that used two polyclonal antibodies specific for PTH 39-84 and PTH 134 was used to measure intact PTH 1-84. The interassay CV was 6.8% at 17 pg/mL and 5.5% at 86 pg/mL. All data are expressed as mean values _+ SEM, except for total triglyceride, apolipoprotein(a), PTH, and insulin sensitivity, for which the results were highly skewed and were normalized by logarithmic transformation and expressed as geometric mean and 95% confidence intervals. A standard computerized statistical program (StatView) was used to perform statistical analysis. The two-tailed, unpaired Student's t4est and chi-squared statistics were used to compare data between subjects on bemodialysis and control subjects, and simple regression analysis was used to examir~eassociations between variables. P < 0.05 was considered to be significant. RESULTS

The clinical characteristics of subjects on hemodialysis and control subjects, and their biochemical and lipid profiles are shown in Tables 1 and 2. The subjects were matched for age, sex, and BMI. Subjects on hemodialysis had a higher systolic blood pressure and a greater proportion

of them were receiving antihypertensive medications, but the difference in the use of antihypertensive medications did not reach statistical significance between the two groups. A n u m b e r of differences were noted in the lipid profile of subjects on hemodialysis compared with control subjects (Table 2). Subjects on hemodialysis had reduced total and L D L cholesterol levels, higher triglyceride and V L D L cholesterol levels, and reduced H D L cholesterol, apolipoprotein AI, LpAI, and L p A I : A I I levels compared with control subjects. The differences in the lipid profile were observed even when only subjects not taking any antihypertensive medication were compared (13 hemodialysis patients and 16 controls). Although subjects on hemodialysis had reduced skinfold thickness, suggesting reduced subcutaneous fat, total body fat, abdominal fat, and percentage of body fat distributed to the abd o m e n on dual-energy x-ray absorptiometry did not differ between the two groups (Table 3). However, the ratio of abdominal fat to lower body fat was higher in subjects on hemodialysis. A b d o m i n a l fat mass was positively correlated with B M I (r 2 = 0.65, P = 0.0005) in all subjects combined. Fasting glucose and insulin were comparable in the two groups of subjects (Table 4), but sub-

58

LEE ET AL

Table 3. Body Fat Quantitation and Distribution in Subjects on Hemodialysis and Control Subjects Hemodialysis Subjects BMI (kg/m 2) Total body fat (g) Abdominal fat (g) Lower body fat (g) Percent of total body fat* Percent of abdominal fat1Percent of lower body fats Skinfold thickness (cm) Suprailiac Subscapular Paraumbilical Waist/hip ratio Abdominal fat/lower body fat

Control Subjects

Probabirity Value

26.1 22,064 2,004 5,562 27.4 10.5 29.2

_+ 4.2 _+ 1491 _+ 210 + 663 ± 2.2 _+ 0.3 _+ 1.2

26.0 18,915 2,163 7,214 31.2 9.7 32.3

± 4.4 _+ 1920 ± 198 _+ 660 _+ 1.7 + 0.5 _+ 1.4

NS NS NS NS NS NS NS

1.3 1.4 1.8 0.93 0.38

_+ 0.1 ± 0.2 _+ 0.1 ± 0.1 _+ 0.02

1.7 2.0 2.1 0.90 0.31

_+ 0.1 _+ 0.1 _+ 0.1 + 0.3 _+ 0.03

<0.05 <0.05 <0.01 NS <0.05

NOTE. Data are given as mean values _+ SEM. * Total body fat/total body tissue. 1- Abdominal fat/total body fat. :~ Lower body fat/total body fat.

jects on hemodialysis had a higher HbAIc that was still within the normal range. Although fasting insulin levels did not differ between the two groups, subjects on hemodialysis had a markedly enhanced first phase insulin response to glucose and reduced sensitivity to the action of insulin (Table 4). Again, these differences were found even when only those subjects not taking any antihypertensive medications were included in the analysis. An inverse relationship was present between insulin sensitivity and first phase insulin

response in each group of subjects (ra = 0.43, P < 0.001 in hemodialysis; r 2 = 0.32, P < 0.005 in control subjects). The difference in insulin sensitivity and triglyceride levels between subjects on hemodialysis and control subjects was due to the relatively low insulin sensitivity and higher triglyceride levels in the leaner individuals on hemodialysis (Fig 2). Using a BMI of 25 (corresponding to an abdominal fat mass of 1,875 g) as the level that distinguished lean from obese subjects, plasma

Table 4. Comparison of Measurements of Glycemia, Insulin Secretion, Insulin Sensitivity, and Nonesterified Fatty Acid Between Hemodialysis and Control Subjects

Fasting glucose (mg/dL) H b A l c (%) Fasting insulin (mU/L) Kg, glucose disappearance rate (min -~) First phase insulin (mUL lmin per mg dL 1) Insulin sensitivity (min -1 per mUL -~ rain) Fasting nonesterified free fatty acid (/~mol/L)

Hemodialysis Subjects

Control Subjects

Probability Value

94 _+ 4 5.1 ± 0.2 9.3 (95% CI, 7.8-11.2)

92 + 4 4.5 ± 0.1 10.2 (95% CI, 8.9-12.0)

NS <0.005 NS

- 5 . 9 x 10 a ( - 5 . 4 x 10 -4)

NS

6.2 × 10 -3 ( - 5 . 0 x 10 -4) 2.72 ± 0.28

1.67 _+ 0.22

2.24 (95% CI, 1.86-2.75)

4.17 (95% CI, 2.95-5.9)

<0.005

480 _+ 60

530 ± 23

NS

NOTE. Data are given as mean values +_ SEM unless indicated otherwise. Abbreviation: CI, confidence interval.

<0.01

DYSLIPIDEMIA IN CHRONIC RENAL FAILURE

59

1.01

0.5 0.4

•

°

0.3

•

°.

o

o

o

o° o

//

1.4 J .

°

1.2 1

pO

o

t% '~

1-

0.2

p<0.OOOS,r2=0.45 i n HD

• %~, 0.1

°

•

°.

°o

/,

0

•

•

0.8-

o

/

,~

p<0 01 r2=0.28 In

/

0,6=

o

*

o

% ,% o © %o% °

o p<0.ooo5,r2=-oA7 in controls ~%,

o

/ p
•*

/ /

°*

-0.1

controls

/

-0.2

,

0.4 ~

.

• %,N

/

/

o ~

0.2-

o,

0

-0.3 1000

2000

3000

4000

1000

5000

100

28

o o

o • ~

~o •

controls

80

in c o n t r o l s

. "~ ~o

5000

p
p<0.005,r2=0.38

•.

4000

HD

27.5

,.<

3000

p=NS,r2=0.04 in

p<0.0005,r2=O.45 in H~

27

2000

ABDOMINAL FAT (g)

ABDOMINAL FAT (g)

25.5

=

o

o o*

6O 50

z0 40

..J

°1

25.5

3O 20

°

25

,°

10

24.0 1000

Z000

3000

4000

5000

ABDOMINAL FAT (g)

0 " 0

1000

2000 3000 ABDOMINAL FAT (g)

4000

5000

Fig 2, Plots of abdominal fat measured by dual-energy x-ray absorptiometry against log triglyceride level, LDL particle diameter, HDL cholesterol level, and log insulin sensitivity in control subjects (open circles) and subjects on hemodialysis (solid diamonds), The solid lines represent subjects on hemodialysis,

triglyceride was higher and insulin sensitivity lower in lean subjects on hemodialysis than in control subjects (Table 5). In subjects with BMI greater than 25, plasma insulin sensitivity and triglyceride did not differ between the two groups. The HDL cholesterol level was lower in subjects on hemodialysis at all levels of abdominal fat. By univariate analysis, abdominal fat mass was a good predictor of triglyceride level, LDL size, and insulin sensitivity in both groups of subjects. Abdominal fat mass was also predictive of HDL cholesterol level in control subjects, but not in subjects on hemodialysis (Fig 2). Further differences were noted between the two groups in the relationship between insulin sensitivity and the lipid profile (Fig 3). In control subjects insulin sensitivity was a significant pre-

dictor of triglyceride, HDL cholesterol levels, and LDL particle size. In contrast, insulin sensitivity in subjects on hemodialysis was not predictive of their lipid parameters. Nonesterified fatty acid levels did not differ between the two groups and bore no correlation with any lipid parameters, abdominal adiposity, or insulin sensitivity. Plasma PTH level in subjects on hemodialysis did not correlate with their insulin sensitivity. A stepwise multiple regression model taking into account the skinfold thicknesses, waist to hip ratio, total body fat mass, abdominal fat mass and percentage of abdominal fat, lower body fat mass and percentage of lower body fat, and abdominal fat to lower body fat mass ratio was used to identify the component of body fat that had the strongest relationship to insulin sensitiv-

60

LEE ET AL

Table 5. Comparison of the Effect of Renal Failure on Insulin Sensitivity and the Lipid Profile of Lean Subjects (Abdominal Fat Mass -<1,875 g} and More Obese Subjects (Abdominal Fat Mass >1,875 9) 1,875

No. Triglyceride (mg/ dL) Insulin sensitivity (min -1 per mUL -1 min) LDL diameter (nm) HDL cholesterol (mg/dL)

Hemodialysis Subjects

Control Subjects

11 120 (95% Cl, 106-158)

9 66 (95% CI, 49-87)

3.09 (95% CI, 2.57-3.72)

> 1,8759 Probability Value

Hemodialysis Subjects

Control Subjects

Probability Value

<0.001

15 137 (95% Cl, 106-177)

13 134 (95% CI, 112-165)

NS

7.41 (95% CI, 5.62-10.0)

<0.005

1.82 (95% CI, 1.45-2.34)

2.75 (95% CI, 2.29-3.31)

NS

26.5 _+ 0.1

26.7 _+ 0.2

NS

25.8 _+ 0.2

37 _+ 4

56 _+ 7

<0.05

34 _+ 4

26 _+ 0.2 45 + 4

NS <0.05

NOTE. Data are given as mean values _+ SEM unless indicated otherwise. Abbreviation: CI, confidence interval.

ity and triglyceride in each group of subjects. The subscapular skinfold thickness was the most significant predictor of insulin sensitivity in hemodialysis subjects, accounting for 65% of the variation in insulin sensitivity (r = 0.81). In contrast, abdominal adiposity was the most significant predictor of insulin sensitivity in control subjects, accounting for 44% of the variation in insulin sensitivity (r = 0.66). Abdominal fat mass also was the most significant predictor of triglyceride in both control subjects and those on hemodialysis (r 2 = 0.29 in hemodialysis, r 2 = 0.39 in control subjects). When the regression analysis included insulin sensitivity as a predictor as well, insulin sensitivity was found to be the most significant predictor of plasma triglyceride in control subjects (r 2 = 0.41), whereas in subjects on hemodialysis, abdominal fat remained the most significant predictor of plasma triglyceride level (r 2 = 0.29). DISCUSSION

Using the IVGTT, we were able to confirm previous observations made with the insulin clamp technique that sensitivity to the action of insulin is reduced in chronic renal failure. 14'33'34 The presence of an enhanced insulin response indicated that pancreatic B-cell function was intact in subjects on hemodialysis, so that the secretion of insulin was able to increase in the face

of peripheral resistance to its action and restore blood glucose level at the same rate as control subjects following injection of the glucose bolus. Previous studies of glucose tolerance in renal failure also have demonstrated an enhanced insulin response to glucose loads in some uremic subjects. 35 The relatively elevated HbAIc in subjects on hemodialysis is not likely to be due to impairment of glucose metabolism, but probably resulted from coelution of carbamylated hemoglobin with glycosylated hemoglobin. 36 Clinical and animal studies have suggested that secondary hyperparathyroidism plays a part in the genesis of glucose intolerance in renal f a i l u r e . 3739 However, in this cross-sectional study, a correlation between the mildly elevated PTH level and insulin sensitivity was not demonstrated. Previous studies of insulin action in subjects with end-stage renal failure did not explore the role of body fat distribution in the relationship between insulin resistance and the lipid profile. The association of abdominal obesity, 24'25particularly visceral obesity, 4° with hypertriglyceridemia and reduced HDL cholesterol is well documented in subjects without renal failure. Similarly, abdominal obesity is correlated with reduced insulin action. 41-44 Our results show that abdominal adiposity is also associated with an unfavorable metabolic profile in renal failure. The presence of renal failure was responsible for insulin resistance and hypertriglyceridemia in the

VS LOG INS SENS (2)

0.4 ..---o

0.5 p=NS,r2=O.12 0.4

% 0,3

p
\ 0

\ 0.3

%

0.2

\ \

.2

\ 0 0.1

)0.I

o

o

%o % \

0

o

o

\

O:

% %

-0.1

-0.t i

% %

-0.2

\ o

-0.2

'%0

e ..

-0.3 0.2

0.4

0.6

0.8

-0.3

LOG INSULIN SENSITIVITY

o 1

0~5

1.5

LOG INSULIN SENSITIVITY

27.5

28

p
p=NS,r2=0.13 o

27.5

27

o

o

27 o

o

"~=26°5

o

..*

2s.s 4

,.?..'"-o

o

/

s-

o

o

o

o

25,5 o o

2S t 25

Z4.S r 0

0.2

0.4

0,6

0,8

0.5

L O G INSULIN SENSITIVITY

1.5

1

LOG INSULIN SENSITIVITY

100 ,

80

O O

70 i==NS,r2=0.18

p<0.05,r2=0,23

8O

60 o

..-

50 o

g~ 50

tO

o

ZO

o ~ 3O

o

P

o

o

o

o o

o

Z0 20 10

10

0 0

0.2

0.4

~

r

0.6

0.8

L O G INSULIN SENSITIVITY

HD

0

0.5

1

t.5

L O G INSULIN SENSITIVITY

CONTROLS

Fig 3. Correlations of log insulin sensitivity with log triglyceride, HDL cholesterol level, and LDL particle size in control subjects and subjects on hemodialysis. Subjects on hemodialysis are represented in the left panels, and control subjects are represented in the right panels.

62

leaner subjects on hemodialysis. In the more obese subjects, abdominal adiposity was a more potent factor than renal failure as a cause of insulin resistance and hypertriglyceridemia. As the level of abdominal adiposity rose, insulin sensitivity declined; the triglyceride level rose at a steeper rate in control subjects, so that renal failure was no longer associated with additional insulin resistance and hypertriglyceridemia. The metabolic properties of abdominal fat differ from peripheral fat in several ways that can account for the impact of abdominal obesity on the metabolic profile. Visceral fat is more sensitive to lipolytic stimuli than subcutaneous fat deposits, 45-47 and as a result of a lower density of insulin receptors, 48 it also is less sensitive to the inhibitory action of insulin on lipolysis. 49 Consequently, visceral fat has a higher turnover than other adipose tissues, 5° which could result in greater portal concentrations of NEFA with three further consequences. First, it could result in an increased hepatic production of triglyceride-rich VLDL particles, as NEFA delivery is probably the rate-limiting factor in hepatic synthesis of VLDL. 51 Second, it could result in elevated systemic NEFA levels and interference with insulin action in muscles by the mechanisms described in relation to the glucose-fatty acid cycle. 52 Third, it could reduce the normal high clearance of insulin on its first pass through the liver. 51 The end result of visceral obesity is reduced insulin sensitivity on the one hand and hypertriglyceridemia on the other. In this study, total abdominal adiposity measured by dual-energy x-ray absorptiometry was the component of body fat that predicted best the metabolic profile in control subjects, even though it does not distinguish intra-abdominal from subcutaneous adiposity. The wide range of abdominal adiposity displayed by this group of control subjects highlights that the role of abdominal adiposity in determining the metabolic profile is continuous and not confined to obese subjects. In subjects on hemodialysis, other factors associated with renal failure also influenced the metabolic profile, so that insulin resistance and hypertriglyceridemia occurred at lower levels of abdominal adiposity than in control subjects. The comparable levels of NEFA in the subjects with renal failure compared with control subjects in

LEE El- AL

this study and in previous studies 34 suggested that an increased turnover of abdominal fat may not be the additional factor responsible for insulin resistance and higher triglyceride levels in renal failure, although portal vein levels of NEFA may have been elevated. Hypertriglyceridemia in chronic renal failure is due predominantly to impaired catabolism of the triglyceride-rich lipoproteins 53-55 as a result of defective action of the postheparin lipases.17'56 Insulin resistance is thought to be an important explanation for these abnormalities in lipoprotein metabolism in renal failure. I7'18 However, none of the previous studies actually measured insulin sensitivity and assessed its contribution to the lipid profile. In the present study, although subjects on hemodialysis were insulin resistant, their LDL particle size and triglyceride level did not correlate with their sensitivity to insulin, as measured by IVGTT. Reduced HDL cholesterol is another feature of insulin resistance. 57 In subjects on hemodialysis, HDL cholesterol level was uniformly low, and it did not correlate with either insulin sensitivity or abdominal adiposity. This finding is in contrast to the correlation between insulin sensitivity and HDL cholesterol level in control subjects, and suggests that the mechanisms regulating HDL cholesterol levels are different in renal failure. There are two possible explanations for these differences between subjects on hemodialysis and control subjects, whose insulin sensitivity was actually a stronger predictor of their lipid profile than abdominal adiposity. First, insulin resistance is not the only factor responsible for defective action of lipoprotein lipase in renal failure. Nondialyzable inhibitors of lipoprotein lipase are present in uremia, 58'59 and chemical modifications of the lipases analogous to carbamylation of hemoglobin 36 also may affect lipase function. A reduced ratio of apoCII, a cofactor for lipoprotein lipase catalyzed lipolysis, and apoCIII, an inhibitor of this process, in plasma 6° and in the triglyceride-rich lipoproteins 61'62of patients with chronic renal failure may further interfere with lipolysis. Second, there are many other abnormalities of lipoprotein metabolism in renal failure, 6~'67 which can also modify the lipid profile of subjects on hemodialysis. Insulin resistance and the profile of high tri-

DYSLIPIDEMIA IN CHRONIC RENAL FAILURE

g l y c e r i d e a n d l o w H D L cholesterol levels f o r m part o f a cluster of m e t a b o l i c a b n o r m a l i t i e s coll e c t i v e l y k n o w n as the m e t a b o l i c s y n d r o m e or s y n d r o m e X. 6s A l l c o m p o n e n t s o f this s y n d r o m e h a v e b e e n s h o w n to b e c a r d i o v a s c u l a r risk factors, a n d i n s u l i n r e s i s t a n c e has b e e n s u g g e s t e d to b e the c o m m o n l i n k b e t w e e n them. 6s'69 S o m e , b u t n o t all, o f the features o f this s y n d r o m e are present i n subjects o n h e m o d i a l y s i s , b u t i n s u l i n resistance does n o t directly a c c o u n t for their a b n o r m a l lipid profile. A b d o m i n a l a d i p o s i t y has a n e g a t i v e i m p a c t o n the m e t a b o l i c profile in subjects o n h e m o d i a l y s i s , b u t the p r e s e n c e o f r e n a l failure itself p r o d u c e s i n s u l i n r e s i s t a n c e in l e a n e r subjects a n d d y s l i p i d e m i a i n all subjects o n h e m o d i alysis c o m p a r e d w i t h control subjects w i t h c o m p a r a b l e levels o f a b d o m i n a l adiposity. D y s l i p i d e m i a a n d i n s u l i n r e s i s t a n c e are i m p o r t a n t c a r d i o v a s c u l a r risk factors, a n d i d e n t i f i c a t i o n o f their d e t e r m i n a n t s i n r e n a l failure c o u l d lead to f o r m u l a t i o n o f strategies for their m o d i f i c a t i o n . ACKNOWLEDGMENT

The authors thank John Santamaria for statistical advice. The Department of Endocrinology, St Vincent's Hospital, performed the insulin and nonesterified free fatty acid assays. Routine biochemical assays were performed by the Department of Chemical Pathology, St Vincent's Hospital. REFERENCES

1. Attman P-O, Alaupovic P, Gustafson A: Serum apolipoprotein profile of patients with chronic renal failure. Kidney Int 32:368-375, 1987 2. Attman P-O, Gustafson A, Alaupovic P, Wang CS: Lipid metabolism in patients with chronic renal failure in the predialytic phase. Contrib Nephrol 65:24-32, 1988 3. Avram MM, Goldwasser P, Burrell D, Antigneni A, Fein P, Mittman N: The uremic dyslipidemia: A cross sectional and longitudinal study. Am J Kidney Dis 20:324-335, 1992 4. Ma KW, Greene EL, Raij L: Cardiovascular risk factors in chronic renal failure and hemodialysis population. Am J Kidney Dis 19:505-513, 1992 5. Bagdade JD, Porte D Jr, Bierman EL: Hypertriglyceridemia: A metabolic consequence of chronic renal failure. N Engl J Med 279:181-185, 1968 6. Brunzell JD, Albers JJ, Haas LB, Goldberg AP, Agadoa L, Sherrard DJ: Prevalence of serum lipid abnormalities in chronic hemodialysis. Metabolism 26:903-910, 1977 7. Cheung AK, Wu LL, Kablitz C, Leypoldt J: Atherogenic lipids and lipoproteins in hemodialysis patients. Am J Kidney Dis 22:271-276, 1993 8. Rubies-Prat J, Espinel E, Joven J, Ras MR, Pira L: High density lipoprotein subfractions in chronic uremia. Am J Kidney Dis 9:60-65, 1987

63

9. Senti M, Romero R, Pedro-Botet J, Pelegri A, Nogues X, Rubies-Prat J: Lipoprotein abnormalities in hyperlipidemic and normolipidemic men on hemodialysis with chronic renal failure. Kidney Int 41:1394-1399, 1992 10. Bergesio F, Monzani G, Ciuti R, Serrato A, Benucci A, Frizzi V, Salvador M: Lipids and apolipoproteins change during the progression of chronic renal failure. Clin Nephrol 38:264-270, 1992 11. O'Neal D, Lee P, Murphy B, Best JD: Low density lipoprotein (LDL) particle size distribution in end stage renal failure treated with hemodialysis or continuous ambulatory peritoneal dialysis. Am J Kidney Dis 27:84-91, 1996 12. Austin MA, Breslow JL, Hennekens CH, Buring JE, Willet WC, Krauss RM: Low density lipoprotein subclass pattern and the risk of myocardial infarction. JAMA 260:1917-1921, 1988 13. Reaven GM, Chen Y-DI, Jeppesen J, Maheux P, Krauss RM: Insulin resistance and hyperinsulinemia in individuals with small, dense, low density lipoprotein particles. J Clin Invest 92:141-146, 1993 14. Schmitz O, Alberti K, Christensen N, Hasling C, Hjollund E, Beck-Nielsen H, Ovskov H: Insulin resistance in glucose homeostasis in uremia as assessed by the hyperinsulinemic euglycaemic clamp technique. Metabolism 34:465473, 1985 15. DeFronzo RA, Smith D, Alvestrand A: Insulin action in uremia. Kidney Int 16:S102-104, 1983 16. Frayn KN: Insulin resistance and lipid metabolism. Curr Opin Lipidol 4:197-204, 1993 17. Chan MK, Persaud J, Varghese Z, Moorhead JF: Pathogenetic roles of post-heparin lipases in lipid abnormalities in hemodialysis patients. Kidney Int 25:812-818, 1984 18. Attman P-O, Samuetsson O, Alaupovic P: Lipoprotein metabolism and renal failure. Am J Kidney Dis 21:573-592, 1993 19. Feinleib M: Epidemiology of obesity in relation to health hazards. Ann Intern Med 103:1019-1024, 1985 20. Fujioka S, Matsuzawa Y, Tokunaga K, Tarni S: Contribution of intraabdominal fat accumulation to the impairment of glucose and lipid metabolism in human obesity. Metabolism 36:54-59, 1987 21. Lapidus L, Bengtsson C, Larsson B, Pennert K, Rybo E, Sjostrom L: Distribution of adipose tissue and risk of cardiovascular disease and death: A 12 year follow up of participants in the population study of women in Gothenberg, Sweden. BMJ 289:1257-1261, 1984 22. Despres JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C: Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arteriosclerosis 10:497-511, 1990 23. Donahue RP, Abbot RD, Bloom E, Reed DM, Yano L: Central adiposity and coronary heart disease risk in men. Lancet 1:822-824, 1987 24. Anderson AJ, Sobocinski KA, Freedman S, Barboriak JJ, Rimm AA, Grnchow HW: Body fat distribution, plasma lipids, and lipoproteins. Arteriosclerosis 8:88-94, 1988 25. Haffner SM, Fong D, Hazuda HP, Pugh JA, Patterson JK: Hyperinsulinemia,upper body adiposity and cardiovascular risk factors in non-diabetics. Metab Clin Exp 37:338-345, 1988

64

26. Peiris AN, Struve MF, Mueller RA, Lee MB: Glucose metabolism in obesity: Influence of body fat distribution. J Clin Endocrinol Metab 67:760-767, 1988 27 . Galvin P, Ward G, Waiters J, Pestell R, Koschmann M, Vaag A, Martin I, Best JD, Alford F: A simple method for quantitation of insulin sensitivity and insulin release from an intravenous glucose tolerance test. Diabetic Med 9:921928, 1992 28. Albano J, Ekins R, Martiz G, Turner R: A sensitive precise radioimmunoassay of serum insulin relying on charcoal separation of bound and free moieties. Acta Endocrinol (Copenh) 70:487-509, 1972 29. Martin IK, Weber KM, Boston RC, Alford FP, Best JD: The effects of epinephrine infusion on determinants of intravenous glucose tolerance in dogs. Am J Physiol 18:E668E673, 1988 30. Goldstein DE: Recent advances in glycosylated hemoglobin measurements. Crit Rev Clin Lab Sci 21:187-228, 1984 31. Chung BH, Segrest JP, Ray MJ, Brunzell JD, Hokanson JE, Krauss RM, Beaudrie K, Cone JT: Single vertical spin density gradient ultracentrifugation, in Segrest JP, Albers JJ (eds): Methods in Enzymology. San Diego, CA, Academic, 1986, pp 181-209 32. Kranss RM, Burke DJ: Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res 23:97-104, 1982 33. DeFronzo RA, Alvestrand A, Smith D, Hendler R, Hendler E, Wahren J: Insulin resistance in uremia. J Clin Invest 67:563-568, 1981 34. Schmitz O, Alberti K, Christensen N, Hasling E, Hjollund E, Beck-Nielsen H, Orskov H: Aspects of glucose homeostasis in uremia as assessed by the hyperinsulinemic euglycemic clamp technique. Metabolism 34:465-473, 1985 35. DeFronzo RA, Andres P, Edgar P, Walker GW: Carbohydrate metabolism in uremia: A review. Medicine 52:469481, 1973 36. Oimomi M, Ishikawa K, Kawasaki T, Kubota S, Yoshimura Y, Baba S: Carbamylation of hemoglobin in renal failure and clinical aspects. Metabolism 33:999-1002, 1984 37. Akmal M, Massry SG, Goldstein DA, Fanti P, Weisz A, Defronzo R: Role of parathyroid hormone in the glucose intolerance of chronic renal failure. J Clin Invest 75:10371044, 1985 38. Mak RHK: Amelioration of hypertension and insulin resistance by 1,25 dihydi-oxycholecalciferol in hemodialysis patients. Pediatr Nephrol 6:377-382, 1992 39. Shoji T, Nishizawa Y, Nishitani H, Yamakawa M, Morii H: Impaired metabolism of high density lipoprotein in uremic patients. Kidney Int 41:1653-1661, 1992 40. Fujioka S, Matsuzawa Y, Tokunaga K, Kawamoto T, Kobatake T, Keno Y, Kotani K, Yoshida S, Tarui S: Improvement of glucose and lipid metabolism associated with selective reduction of intraabdominal visceral fat in premenopansal women with visceral obesity. Int J Obes 15:853-859, 1991 41. Evans DJ, Hoffman RG, Kalkhoff RK, Kissebah AH: Relationship of body fat topography to insulin sensitivity and metabolic profiles in premenopausal women. Metabolism 33:68-75, 1984

LEE ET AL

42. Krotldewski M, Bjorntorp P, Sjostrom L, Smith U: Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest 72:1150-1162, 1983 43. Seidell JC, Bjorntorp P, Sjostrom L, Kvist H, Sannerstedt R: Visceral fat accumulation in men is positively associated with insulin, glucose and C-peptide levels, but negatively with testosterone levels. Metab Clin Exp 39:897-901, 1990 44. Pouliot MC, Despres JP, Nadeau A, Moorjani S, Prud'Homme D, Lupien PJ, Tremblay A, Bouchard C: Visceral obesity in men: Associations with glucose intolerance, plasma insulin, and lipoprotein levels. Diabetes 41:826-834, 1992 45. Rebuffe-Scrive M, Anderson B, Olbe L, Bjorntorp P: Metabolism of adipose tissue of intraabdominal depots of non-obese men and women. Metabolism 38:453-461, 1989 46. Hellmer J, Marcus C, Sonnenfeld T, Arner P: Mechanisms for differences in lipolysis between human subcutaneous and omental fat cells. J Clin Endocrinol Metab 75:1520, 1992 47. Arner P, Hellstrom L, Wahrenberg H, Bronengard M: Beta-adrenergic receptor expression in human fat cells from different regions. J Clin Invest 86:1595-1600, 1990 48. Bolinder J, Kager L, Ostman J, Arner P: Differences at the receptor and postreceptor levels between human omental and subcutaneous adipose tissue in the action of insulin on lipolysis. Diabetes 32:117-122, 1983 49. Bolinder J, Engfeldt P, Ostman J, Amer P: Site differences in insulin receptor binding and insulin action in subcutaneous fat of obese females. J Clin Endocrinol Metab 57:455-459, 1983 50. Bjorntorp P: Metabolic implications of body fat distribution. Diabetes Care 14:1132-1143, 1991 51. Bjorntorp P: "Portal" adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis 10:493-497, 1990 52. Randle PJ, Garland PB, Hales CN, Newsholme EA: The glucose fatty acid cycle: Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 2:785-789, 1963 53. Cattran DC, Fenton SSA, Wilson DR, Steiner G: Defective triglyceride removal in lipemia associated with peritoneal dialysis and hemodialysis. Ann Intern Med 85:29-33, 1976 54. Savdie E, Gibson JC, Crawford GA, Simons LA, Mahony JF: Impaired triglyceride clearance as a feature of both uremic and posttransplant triglyceridemia. Kidney Int 18:774-782, 1980 55. Roullett J-B, Lacour B, Yvert JP, Prat JJ, Drueke T: Factors in increase of triglyceride rich lipoproteins in uremic rats. Kidney Int 27:420-425, 1985 56. Applebanm-Bowden D, Goldberg AP, Hazzard WR, Sherrard D J, Brunzell JD, Huttunen JK, Nikkila EA, Ehnholm C: Posthepmin plasma triglyceride lipases in chronic hemodialysis: Evidence for a role for hepatic lipase in lipoprotein metabolism. Metabolism 28:917-924, 1979 57. Karhapaa P, Malkki M, Laakso M: Isolated low HDL cholesterol. Diabetes 43:411-417, 1994 58. Murase T, Cattran DC, Rubenstein B, Steiner G: Inhi-

DYSLIPIDEMIA IN CHRONIC RENAL FAILURE bition of lipoprotein lipase by uremic plasma, a possible cause of hypertriglyceridemia. Metabolism 24:1279-1286, 1975 59. Crawford GA, Mahony JF, Stewart JH: Impaired lipoprotein lipase activation by uremic and posttransplant sera. Clin Sci 60:73-80, 1981 60. Attman PO, Alaupovic P: Lipid and apolipoprotein profiles of uremic dyslipoproteinemia-relation to renal function and dialysis. Nephron 57:401-410, 1991 61. Wakabayashi I, Okubo M, Shimada H, Sato N, Koide A, Marumo F, Hakamura H: Decreased VLDL apoprotein CII/apoprotein CIII ratio may be seen in both normotriglyceridemic and hypertriglyceridemic patients on chronic hemodialysis treatment. Metabolism 36:815-820, 1987 62. Alsayed N, Rebourcet R: Abnormal concentrations of CII, CIII, and E apolipoproteins among apolipoprotein B containing, B free, and AI containing lipoprotein particles in hemodialysis patients. Clin Chem 37:387-393, 1991 63. Guarnieri G, Moracchiello M, Campanacci L, Ursini F, Ferri L, Valente M, Gregolin C: Lecithin cholesterol acyltransferase (LCAT) activity in chronic uremia. Kidney Int 13:$26-$30, 1978 (suppl 8)

65 64. Dieplinger H, Schoenfeld PY, Fielding CJ: Plasma cholesterol metabolism in end stage renal disease--Difference between treatment by hemodialysis or peritoneal dialysis. J Clin Invest 77:1071-1083, 1986 65. Corboy J, Sutherland WH, Walker RJ, Robertson MC, Cox CM: Cholesteryl ester transfer in patients with renal failure or renal transplant. Kidney Int 46:1147-1153, 1994 66. Weintraub M, Burstein A, Rassin T, Liron M, Ringel Y, Cabili S, Blum M, Peer G, Iaina A: Severe defect in clearing postprandial chylomicron remnants in dialysis patients. Kidney Int 42:1247-1252, 1992 67. Dieplinger H, Lobentanz E-M, Konig H, GrafH, Sandholzer C, Matthys E, Rosseneu M, Utermann G: Plasma apolipoprotein A IV metabolism in patients with chronic renal disease. Eur J Clin Invest 22:166-174, 1992 68. Reaven GM: Role of insulin resistance in human disease. Diabetes 37:1595-1607, 1988 69. Ferrannini E, Haffner SM, Mitchell BD, Stern MP: Hyperinsulinemia: The key feature of a cardiovascular and metabolic syndrome. Diabetologia 34:416-422, 1991