Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1995;92:1770-1778

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Austin, M. A.
Right arrow Articles by Laakso, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Austin, M. A.
Right arrow Articles by Laakso, M.

(Circulation. 1995;92:1770-1778.)
© 1995 American Heart Association, Inc.

Articles

Prospective Study of Small LDLs as a Risk Factor for Non–Insulin Dependent Diabetes Mellitus in Elderly Men and Women

Melissa A. Austin, PhD; Leena Mykkänen, MD; Johanna Kuusisto, MD; Karen L. Edwards, MS; Carrie Nelson, BA; Steven M. Haffner, MD; Kalevi Pyörälä, MD; Markku Laakso, MD

From the Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle (M.A.A., K.L.E., C.N.); the Department of Medicine, University of Kuopio, Finland (L.M., J.K., K.P., M.L.); and the Division of Clinical Epidemiology, Department of Medicine, University of Texas Health Science Center at San Antonio (S.M.H.).

Correspondence to Melissa A. Austin, PhD, Department of Epidemiology, Box 358770, School of Public Health and Community Medicine, University of Washington, Seattle, WA 98195. E-mail maustin@u.washington.edu.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Background The excess risk of atherosclerosis among patients with non–insulin dependent diabetes mellitus (NIDDM) is well documented. However, the presence of conventional risk factors cannot fully account for this excess risk, and the underlying mechanism remains to be elucidated. The present study prospectively evaluated the role of small LDL, a known risk factor for coronary heart disease, as a risk factor for the development of NIDDM.

Methods and Results The study was based on a nested case-control sample of 204 elderly men and women from Kuopio, Finland. LDL subclasses were characterized by size with 2% to 14% polyacrylamide gels produced by recently developed methods. Logistic regression analysis showed that subjects with a predominance of small LDL (LDL subclass phenotype B) had a greater than twofold increased risk for developing NIDDM over the 3.5-year follow-up period. This association was independent of age, sex, glucose intolerance, and body mass index but was not independent of fasting triglyceride or insulin levels. Further, an increase of 5Å in LDL diameter was associated with a 16% decrease in risk of NIDDM, and a composite variable of LDL diameter and triglyceride and HDL cholesterol concentrations, identified by principal-components analysis, was also associated with NIDDM. These associations may be attributable to the role of small LDL as a marker for insulin resistance.

Conclusions This study is the first to demonstrate that a predominance of small LDL particles is a risk factor for the future development of NIDDM, and it implies that small LDL contributes to risk of coronary heart disease in prediabetics.


Key Words: lipoproteins • diabetes mellitus • aging • risk factors


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
The excess risk of atherosclerosis among patients with NIDDM is well documented.1 2 However, the presence of conventional risk factors cannot fully account for this excess, and the underlying mechanism remains to be elucidated. At least three prospective studies have demonstrated that baseline cardiovascular risk factor levels predict incident NIDDM.3 4 5 An 8-year follow-up of Mexican Americans found higher baseline levels of total cholesterol, LDL cholesterol, triglyceride, fasting glucose and insulin, 2-hour glucose, BMI, and blood pressure and lower levels of HDL cholesterol among those who developed NIDDM compared with those who remained nondiabetic.3 Similarly, two studies in older populations, one in southern California and another in Finland, both showed more atherogenic risk factor profiles, including increased triglycerides and blood pressure, among men and women who developed incident NIDDM.4 5 These studies have led to speculation that "the increased atherogenicity of the prediabetic phase may make an important contribution to the subsequent risk of coronary heart disease even before overt diabetes develops."3

A number of case-control studies have recently established that a predominance of small, dense LDL (LDL subclass phenotype B) is associated with risk of coronary heart disease.6 7 Interestingly, other studies have shown that a phenotype B is also an integral feature of the insulin resistance syndrome.8 9 10 11 12 13 14 For example, in a large sample of women, LDL subclass phenotype B has been shown to be independently associated with all of the risk factors that characterize the syndrome.12 In a more direct evaluation using the insulin suppression test, Reaven and colleagues13 demonstrated that subjects with LDL subclass phenotype B were insulin resistant and had increased plasma levels of glucose, insulin, and triglyceride, decreased HDL cholesterol levels, and higher blood pressure. Several other studies have demonstrated the association of small, dense LDL with increased plasma triglyceride and decreased HDL cholesterol6 15 16 17 18 and with obesity and increased waist-to-hip ratio.19 20 21 22

Small, dense LDL has also been associated with NIDDM itself in both women and men.12 23 24 In the same large sample of women as noted above, the prevalence of phenotype B increased from 15% among nondiabetic women to 34% in women with glucose intolerance and to 67% in women with diabetes.12 In a sample of normolipidemic diabetic men, there was a greater than twofold increase in the prevalence of LDL subclass phenotype B compared with age-matched control subjects with similar lipid levels.23 Similarly, among diabetic men, half on diet only and half on diet and sulfonylurea therapy, the density of LDL particles was increased in both groups compared with nondiabetic control subjects.24 Importantly, these studies have all been cross-sectional, and prospective data are needed to establish whether LDL subclass phenotype B precedes the onset of NIDDM.

The purpose of the present study is to prospectively evaluate the role of small, dense LDL as a risk factor for incident NIDDM using a nested case-control sample of elderly men and women from Kuopio, Finland. In addition to standard multivariate logistic regression analysis to assess this association, principal-components analysis is used to create a composite lipoprotein risk factor variable that incorporates the strong correlations between small LDL, triglyceride, and HDL cholesterol into the analysis.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Study Subjects
This prospective analysis is based on a nested case-control subsample from a well-characterized cohort of elderly subjects from Kuopio, Finland.5 25 26 The original cohort, studied at baseline during 1986 through 1988, consisted of 1300 randomly selected men and women 65 to 74 years old from the Kuopio population.5 26 The follow-up visit was conducted during 1990 and 1991, on average 3.5 years after baseline (42±4 months).5 Among the 1192 subjects living at follow-up, the response rate was 88%, for a total of 1054 participants. The study was approved by the Kuopio University Hospital, and all subjects gave informed consent.

For the nested case-control design, only those subjects with a complete OGTT, without a physician's diagnosis of NIDDM, and without diabetic OGTT at baseline were eligible. At follow-up, the WHO diagnostic criteria for NIDDM were used,27 based on a 75-g OGTT.5 The 69 incident cases of NIDDM identified in this manner included 30 men and 39 women, who were selected as the case patients for this analysis.

Two groups of control subjects, all of whom were free of diabetes at both baseline and follow-up, were selected. Control group 1 was matched to cases for sex only. Control group 2 was more extensively matched to cases for sex, glucose intolerance status, and BMI at baseline. Control groups 1 and 2 thus consisted of 69 individuals each, for a total of 138 control subjects.

Baseline Data Collection
Details of baseline data collection have been described in detail previously.5 For this analysis, BMI was used as an index of overall obesity and was calculated as weight in kilograms divided by height in meters squared, and waist-to-hip ratio was used as a measure of body fat distribution.5 Family history of diabetes was defined as positive if one or more of the parents or siblings of the subject was reported to have NIDDM. Drug treatment for hypertension was also recorded for all study subjects. Glucose intolerance at baseline was defined according to WHO criteria based on OGTT results.27

Serum lipid and insulin levels were determined on fresh samples and samples frozen at -70°C, respectively, after a 12-hour overnight fast, as previously described.5 Serum samples used for characterizing LDL subclasses in the cases and control subjects were also collected at baseline after a 12-hour fast and stored at -70°C for an average of 6.4 years (67 to 92 months). They were transported to Seattle on dry ice and were never thawed until gradient gel electrophoresis procedures were performed. Samples were not available on 1 case and corresponding control subjects. Thus, the final sample size was 68 cases and 136 control subjects.

Characterization of LDL Subclasses
Gel Production
LDL subclasses were characterized by use of 2% to 14% polyacrylamide gels and electrophoresis procedures originally described by Krauss and Burke28 and Nichols et al.29 The gels were produced in one of our laboratories (M.A.A.) by the method recently described by Rainwater and colleagues30 for producing 4% to 30% gels, with modifications to create a 2% to 14% polyacrylamide gradient. To provide reproducible flow rates and uniform pouring conditions, a commercial dual-pump gradient controller and software package (Chemresearch Data Management Controller, ISCO) were used to interface a PC computer with peristaltic pumps (Tris pumps, ISCO).

Glass gel cassettes assembled in advance were used to pour gels in batches of eight with a Pharmacia GSC-8 casting chamber kept at 20°C to 22°C and leveled. Polyacrylamide (Bio Rad Laboratories) stock solutions of 2% (wt/vol, 1.65% total, 5.6% cross-linker) and 14% (wt/vol, 14% total, 4% cross-linker) containing 5% (wt/vol) sucrose were filtered and stored at 4°C. Just before pouring, freshly prepared 10% (wt/vol) ammonium persulfate (Bio Rad) and DAMPN (Sigma Chemical Co) were added. The 2% acrylamide casting solution contained 1 vol 2% acrylamide stock solution, 0.008 vol ammonium persulfate, and 0.0005 vol DAMPN. The 14% casting solution contained 1 vol 14% acrylamide stock solution, 0.0003 vol ammonium persulfate, and 0.0005 vol DAMPN.

At rates controlled by the dual-pump linear gradient casting program, the casting solutions were pumped first into a mixing chamber and then into the casting chamber under a layer of ethanol/water 1:4 (vol/vol). When the gel pouring was completed, a 60% sucrose solution was pumped through the system to displace the polymerization solution upward into the cassettes. The gels were left undisturbed for 3 hours to polymerize, immersed in distilled H2O for 5 minutes, and then stored at 4°C in Ziploc bags with 2 mL electrophoresis buffer.

Electrophoresis
Gradient gel electrophoresis was performed with the investigators blinded to the case-control status of the serum samples. Each case and the two corresponding matched controls were run in adjacent lanes on the same gel, and two such "triplets" were run on each gel. Two gels were electrophoresed for each triplet: one using whole serum stained for lipid with oil red O (Aldrich No. 19, 819-6) and another using LDL (d<1.063 g/mL) isolated from the same serum aliquot and stained for protein with Coomassie blue R250 (Sigma No. B-0630). A set of high-molecular-weight standards was also run on each gel and used to construct a quadratic calibration curve for estimating the diameter of LDL subclasses, as previously described.28 In addition, two quality control samples with well-characterized LDL subclass phenotypes were run on each gel. Each gel lane was scanned with a QuickScan densitometer (Helena Laboratories) interfaced with a PC computer via an analog-to-digital converter and the Hoefer GS365 data system.

Data Processing and LDL Subclass Phenotype Determination
Gradient gel electrophoresis data were processed with software developed in our laboratory (Gel Scan Calibration Program). First, the peaks and shoulders corresponding to LDL subclasses were marked and denoted as "major," "minor," or "insignificant." The calibration curve determined from the high-molecular-weight standards was then applied to each peak to estimate the LDL particle diameter. The estimated diameter for the major peak in each scan was called the peak particle diameter (LDL-PPD) and was used as a continuous variable in statistical analyses. The diameters of the major peaks of the quality control samples were also estimated.

The resulting scan data for each sample lane were then presented to each of three evaluators who independently classified the LDL subclass phenotype. Since each subject had results from two gels (whole serum and isolated LDL), a total of six LDL subclass phenotype classifications were made for each subject. These classifications were performed with the investigators blinded both to case-control status and to the data from the other gel. If all phenotype classifications were unanimous, the appropriate phenotype was assigned. If not, the evaluators met to determine a consensus phenotype or the gel was rerun, as appropriate. The LDL subclass phenotype definitions used were (1) phenotype A: major LDL peak between 260Å and 280Å, with skewing of the gel scan to the right; (2) phenotype B: major LDL peak between 240Å and 255Å, with skewing of the gel scan to the left; and (3) phenotype I (intermediate): major peak between 255Å and 260Å, with either a symmetrical scan or skewing to the left or double major peaks in this range.

For statistical analyses requiring a dichotomous variable, analyses were repeated using both the "narrow" and "broad" definitions of phenotype B.31 In the narrow definition, phenotype I subjects are grouped with those with phenotype A, whereas in the broad definition, those with phenotype I are grouped with phenotype B.

Comparison With Pharmacia Gels
Before production ceased, 2% to 16% Pharmacia gels were used in most laboratories for characterization of LDL subclasses.28 31 The gels produced in our laboratory at the University of Washington were compared with Pharmacia gels by use of LDL-PPD values estimated for two quality control samples. These quality control samples were aliquots from a single plasma sample obtained from each of two well-characterized study subjects, one with LDL subclass phenotype A (quality control sample 1) and one with phenotype B (quality control sample 2). Based on 24 Pharmacia gels, the mean LDL-PPD values for the quality control samples were 264.9±4.9Å (mean±SD) and 248.3±5.2Å, respectively. A total of 122 gels in our laboratory have been produced to date with nearly identical mean LDL-PPD mean values for the quality control samples (264.3±3.2 and 246.6±4.5Å, respectively, P>.20 compared with Pharmacia gels). The subset of these gels used for the present study also had very similar LDL-PPD mean values (263.5±2.7 and 247.5±3.4Å, P>.20 compared with Pharmacia gels). In fact, the SD values, reflecting intergel variation, were lower in the University of Washington gels than the Pharmacia gels. Thus, the LDL subclass characterization using gels produced in our laboratory appears comparable to those using Pharmacia gels.

Statistical Analysis
Comparisons of mean values between groups were made with either Student's t test or one-way ANOVA with orthogonal contrasts for multiple comparisons.32 33 Analysis of categorical variables was performed with {chi}2 tests with appropriate degrees of freedom. Relations between risk factor levels were examined with Pearson's product moment correlation coefficients. Because the distributions of triglyceride and of fasting insulin were highly skewed (skewness, 2.45 and 1.31, respectively, for all study subjects), the natural logarithms of these variables were used in all calculations. However, mean values are reported in antilogarithmic units for ease of interpretation.

The association between LDL subclasses at baseline and incident NIDDM during the follow-up period was assessed by ORs estimated by logistic regression, including the matching variables.34 In this nested case-control study design, these ORs are estimates of the relative risks of developing NIDDM among those subjects with phenotype B at baseline compared with those with phenotype A at baseline. Because the number of prediabetic cases in the study was relatively small and because the frequency of phenotype B was not significantly different in the two control groups ({chi}2=1.47, P=.23 in the narrow definition of phenotype B), the control groups were combined to increase the statistical power of the analysis. Multivariate analyses adjusting for additional covariates were also performed using unconditional logistic regression. All analyses were repeated using the narrow definition of LDL subclass phenotype B, the broad definition of phenotype B, and LDL-PPD from the isolated LDL gels. Interaction terms involving triglyceride, fasting insulin, and LDL subclass phenotype B were also evaluated in the multivariate models, but no significant interactions were detected.

Because LDL-PPD, triglyceride, and HDL cholesterol levels were found to be highly correlated, principal-components analysis was used to reduce multicollinearity between these risk factors by defining a composite lipoprotein risk factor variable. These procedures have recently been described in detail in an analysis of the insulin resistance syndrome.35 Briefly, principal-components analysis transforms a set of correlated risk factors to a linear combination of the variables that accounts for the maximum proportion of the total variance.36 37 Factor loadings then describe the correlation between the newly defined component and the original risk factor variables. Standardized scoring coefficients are also estimated and used to compute a factor score for each individual subject. This factor score is a linear combination of the variables and is calculated on the basis of risk factor levels standardized to have a mean of zero and a variance of {approx}1.0.36 A factor score was calculated for all study subjects and used as a continuous variable in the logistic regression analysis, as described above.

All computations were performed with the STATISTICAL ANALYSIS SYSTEM (SAS).38


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Characteristics of Cases and Control Subjects at Baseline
The baseline characteristics of study subjects are summarized in Table 1Down for prediabetic cases and for each of the two control groups. The 68 cases who developed NIDDM during the follow-up period included 29 men and 39 women. Because of the matching criteria, each of the two control groups contained the same number of men and women. The mean ages of all three groups were nearly identical (69 years). Glucose intolerance was present in more than two thirds of the cases at baseline; the prevalence was much lower in control group 1 (17%) and was similar to cases in control group 2, again because of the matching criteria. In addition, mean BMI was virtually identical in the NIDDM cases and in control group 2 because of matching but was slightly lower in control group 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Baseline Characteristics of Prediabetic NIDDM Cases and Control Groups (1986-1988)

Both family history of diabetes and use of hypertension medication were somewhat increased in cases compared with the control groups, but these differences were not statistically significant. Of those subjects using hypertension medication, 79.2% of cases and 69.4% of all control subjects were taking ß-blockers (data not shown, P=.404 for cases versus control subjects). Mean values of total cholesterol and HDL cholesterol were nearly the same in cases and control subjects. By ANOVA, mean triglyceride levels were higher in cases than in the two control groups (2.2, 1.6, and 1.9 mmol/L, respectively), but this difference was significant only for control group 1. Further, mean triglyceride values in the two control groups did not differ (P>.05). Fasting insulin levels were also significantly higher in prediabetic cases at baseline compared with each of the control groups (P<.05).

LDL Subclasses and the Development of NIDDM
The associations between LDL subclasses at baseline and the development of NIDDM at follow-up is summarized in Table 2Down, based on the narrow definition of phenotype B and on LDL-PPD. Only 10 subjects (5%) in the study had an intermediate phenotype at baseline: 1 in the case group, 6 in control group 1, and 3 in control group 2. LDL subclass phenotype B was present in 21% of cases, compared with 6% in control group 1 and 12% in control group 2 (Table 2ADown). Although the OR from logistic regression for the association between the development of NIDDM and phenotype B was statistically significant in control group 1 (OR, 4.06; P=.02), after matching for BMI and glucose intolerance, the result did not reach significance in control group 2 (OR, 2.03; P=.15). For the two control groups combined and adjusting for all matching variables, the OR was 2.4 (P=.05; 95% CI, 1.00 to 5.71). However, this relation was not significant in the broad definition of phenotype B for either the individual or combined control groups, with ORs ranging from 1.4 to 1.6 (data not shown). Thus, small, dense LDL was associated with a greater than twofold increased risk of subsequent NIDDM in these older men and women in the narrow definition of phenotype B.


View this table:
[in this window]
[in a new window]
  		Table 2. Association of LDL Subclasses at Baseline and Development of NIDDM at Follow-up

Based on the LDL-PPD estimates from both the whole plasma gels and the isolated LDL gels (Table 2BUp), lower mean values were seen at baseline among the prediabetic cases compared with control subjects. These results were significant both for control group 1 and for the combined control group (P<.05).

Multivariate Logistic Regression Analysis of LDL Subclasses
Table 3Down summarizes the prospective association between risk factors at baseline and incident NIDDM during the follow-up period by reporting ORs and 95% CIs from multivariate logistic regression models. Because the two control groups have been combined for these analyses, all the matching variables (sex, glucose intolerance status, BMI) and age are included in each logistic regression model. Of these, only glucose intolerance was a significant covariate (P<.05 for all models). However, it is important to note that control group 2 was matched to cases on this variable. Other potential confounding variables, including waist-to-hip ratio, family history of diabetes, hypertension medication, lipid levels, and fasting plasma insulin are evaluated as well.


View this table:
[in this window]
[in a new window]
  		Table 3. Association of LDL Subclasses at Baseline and the Development of NIDDM at Follow-up Based on Multivariate Logistic Regression Analysis

Using the narrow definition of LDL subclass phenotype B and including only the matching variables in the model (Table 3AUp, model 1), the ORs for phenotype B and incident NIDDM was 2.4, as previously shown in Table 2AUp. Since the OR for phenotype B was virtually unchanged (2.39 versus 2.42) after adjustment for the potential confounding effects of waist-to-hip ratio, family history of diabetes, and the use of hypertension medication (model 2), these variables are not included in subsequent models. Neither baseline total cholesterol nor HDL cholesterol was prospectively associated with NIDDM, and the OR for phenotype B remained constant when these variables were included in the models (models 3 and 4). However, baseline triglyceride was associated with incident NIDDM (model 5; OR, 2.3 for an increase of 1 natural logarithmic unit of triglyceride; P<=.10). Similarly, baseline fasting insulin was significantly associated with incident NIDDM (model 6). Including either of these variables in the model reduced the OR for phenotype B to a nonsignificant value (1.40 and 1.59, respectively).

The same series of logistic regression models was constructed with the broad definition of phenotype B (data not shown). The relation between phenotype B at baseline by this definition and incident NIDDM was not statistically significant, with ORs ranging from 1.3 to 1.4 (data not shown). Again, both triglyceride and fasting insulin were significantly associated with NIDDM, and the association with phenotype B was eliminated as well.

Using LDL peak particle diameter as a continuous variable resulted in an OR <1.0, since increases in LDL-PPD indicate larger LDL size (Table 3BUp, model 1; OR, 0.84 for 5-Å increase in LDL-PPD; 95% CI, 0.70 to 1.02). However, this association was not statistically significant (P=.06). The magnitude of this OR and the width of the CIs remained similar in models including waist-to-hip ratio, family history of diabetes, use of hypertension medication, total cholesterol, and HDL cholesterol (models 2, 3, and 4). However, no association between LDL size and NIDDM was seen when either triglyceride or fasting insulin was included in the model (models 5 and 6).

Principal-Components Analysis
Several studies have reported strong interrelations between LDL size and/or density and both increased triglyceride and decreased HDL cholesterol,6 15 16 17 18 and it has been proposed that small, dense LDL may be a marker for an abnormality of triglyceride metabolism.39 For this reason, principal-components analysis was used to construct a composite lipoprotein variable to further evaluate relations between lipoprotein risk factors and risk of NIDDM.

The correlations of LDL-PPD, triglyceride, HDL cholesterol, and total cholesterol are shown for cases and control subjects in Table 4ADown. As expected, LDL-PPD was strongly inversely associated with triglyceride among both prediabetic cases and control subjects (r=-.75 and -.61, respectively) and positively associated with HDL cholesterol (r=+.63 and +.52, respectively), indicating that small LDL is associated with increased triglyceride and with decreased HDL cholesterol. Triglyceride was also significantly inversely associated with HDL cholesterol among both prediabetic cases and control subjects. Because correlations between LDL-PPD and total cholesterol were approximately zero, total cholesterol was not included in the principal-components analysis.


View this table:
[in this window]
[in a new window]
  		Table 4. Principal-Components Analysis of Lipoprotein Variables at Baseline

The results of the principal-components analysis based on all study subjects are shown in Table 4BUp. One component that explained 71% of the variance in three variables included in the analysis was found. Consistent with the univariate correlations, the factor loadings for LDL-PPD and for HDL cholesterol were positive and the loading for triglyceride was negative; each was statistically significant by conventional criteria.36 The interpretation of the component reflects increased LDL size and HDL cholesterol and decreased triglyceride and thus could be viewed as a "protective" lipoprotein factor in terms of CHD risk. Virtually the same component was obtained when only control group 1 was used for the principal-components analysis (factor loadings of 0.92, -0.87, and 0.80 for LDL-PPD, triglyceride, and HDL cholesterol, respectively).

Next, a factor score was calculated for each study subject based on the coefficients in Table 4BUp and was used as a composite "lipoprotein factor score" variable in logistic regression (Table 5Down). Model 1 shows that the lipoprotein factor score was inversely associated with incident NIDDM, with a significant OR of 0.71 (95% CI, 0.51 to 0.99; P<=.05), adjusting only for the matching variables. This result indicates a 30% decreased risk of NIDDM associated with a one-unit increase in this lipoprotein factor score at baseline. The magnitude of the OR remained similar with adjustment for waist-to-hip ratio, family history of diabetes, and hypertension medication use (model 2). However, the OR for the lipoprotein score was not significant when fasting insulin level was included in the model (model 3).


View this table:
[in this window]
[in a new window]
  		Table 5. Association of Lipoprotein Factor Score at Baseline and Development of NIDDM at Follow-up Based on Multivariate Logistic Regression Analysis


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
The results of this study in older men and women show that small LDL is prospectively associated with the development of NIDDM. By the narrow definition of LDL subclass phenotype B (in which intermediate phenotypes are grouped with phenotype A subjects), subjects with small, dense LDL had a greater than twofold increased risk of incident NIDDM over the 3.5-year follow-up period independent of age, sex, glucose intolerance, and BMI. Although this association was not significant by the broad definition of phenotype B (in which intermediate phenotypes are grouped with phenotype B), an increase of 5Å in LDL size was associated with a 16% decrease in risk of NIDDM. Previous studies have demonstrated a cross-sectional association between small, dense LDL and diabetes and the insulin resistance syndrome,12 13 14 23 24 but this study is the first to demonstrate that small, dense LDL is a risk factor for NIDDM in prediabetic subjects without overt hyperglycemia. Importantly, the prospective nature of the study confirms that the presence of LDL subclass phenotype B precedes the onset of diabetes in this older age group and implies that small LDL may contribute to CHD risk among prediabetics.

The relation between small LDL and incident NIDDM was maintained in multivariate analysis after waist-to-hip ratio, family history of diabetes, and use of hypertension medication, including ß-blockers, were controlled for. Similar ORs were also obtained after total and HDL cholesterol were controlled for. However, adjusting for either fasting triglyceride or insulin reduced the association of small LDL and NIDDM to nonsignificance (Tables 3AUp and 3BUp). This confounding effect is attributable to the strong correlations between small LDL and both triglyceride and insulin levels, reflecting the clustering of risk factors that characterize the insulin resistance syndrome.8 9 10 11 40 Since the presence of this syndrome is known to be a risk factor for the development of NIDDM, the results presented here imply that the association between LDL subclass phenotype B and NIDDM reflects the role of small, dense LDL as a marker for insulin resistance12 13 rather than a causal relation between this lipoprotein phenotype and the development of diabetes.

A recent review of lipoprotein metabolism in diabetes, however, examined the mechanisms that may underlie the combination of lipoprotein abnormalities found in NIDDM.41 Interestingly, a double-blind study of 16 NIDDM patients treated with gemfibrozil found significant decreases in triglyceride and increases in HDL cholesterol and simultaneous changes in LDL toward larger, less buoyant particles.42 These results show that the lipoprotein abnormalities in NIDDM can be reversed and confirm the strong interrelations between triglycerides, HDL cholesterol, and LDL size and density in diabetes. In the present study, this statistical problem of multicollinearity was addressed directly: a composite lipoprotein variable including LDL-PPD, triglyceride, and HDL cholesterol was obtained by principal-components analysis. The composite variable (lipoprotein factor score) was also significantly related to incidence of NIDDM (Table 5Up), suggesting that these variables may, in combination, reflect a common, underlying pathological mechanism that predisposes to NIDDM.

In previous cross-sectional and prospective studies,43 44 45 46 47 lipoprotein abnormalities have been shown to be associated with CHD among patients with either glucose intolerance or NIDDM. For example, in the Paris Prospective Study, plasma triglyceride predicted CHD death independent of a variety of other risk factors, including plasma cholesterol and fasting and 2-hour post–glucose load insulin levels.45 Among men of Japanese ancestry with abnormal glucose tolerance, low HDL cholesterol was independently associated with incident CHD during 18 years of follow-up in Honolulu.46 In a study of men and women with established NIDDM in Finland, low levels of HDL cholesterol and HDL2 cholesterol and high total triglycerides, VLDL cholesterol, and VLDL triglyceride all predicted CHD events during a 7-year follow-up study.47 Because LDL subclass phenotype B has been shown to be associated with CHD risk in a number of case-control studies,6 48 49 50 51 52 53 54 55 the present study suggests that LDL subclass phenotype B is also a marker for increased risk of CHD among prediabetics.

The results of this study have important clinical and public health implications. The demonstration of lipoprotein disorders preceding the onset of frank diabetes in this elderly population confirms that lipoprotein-related susceptibility to atherosclerosis is already present in the prediabetic state and is less likely to be a consequence of NIDDM itself.3 Furthermore, these findings suggest that intervention to normalize lipoprotein profiles, based on either lifestyle modifications or pharmacological agents, may be more effective in reducing CHD risk if implemented during the prediabetic state rather than after the onset of NIDDM. Such an approach could have the potential to prevent CHD in this significant subgroup of the population.

Interestingly, nondiabetic subjects with a parental history of diabetes have been shown to have a more atherogenic risk factor profile than those without such a parental history,56 including higher plasma triglyceride levels, lower HDL cholesterol levels, and higher BMI, blood pressure, and insulin levels. Similarly, cardiovascular risk factors have been shown to cluster among relatives with a family history of NIDDM.57 A number of studies have also demonstrated that small, dense LDL is genetically influenced.58 For example, recent results from a large study of women twins have demonstrated that one third to one half of the variation in LDL size is attributable to genetic influences, a highly statistically significant result.59 Family studies using complex segregation analysis have established that the inheritance of small, dense LDL is consistent with a single major gene effect, and these results are further supported by a report of linkage between small, dense LDL and markers on the short arm of chromosome 19.58 60 Thus, it is tempting to speculate that the genetic influences on small, dense LDL are involved in this familial aggregation of CHD risk factors in NIDDM.

In conclusion, this prospective study has demonstrated that small, dense LDL is associated with a greater than twofold increased risk for NIDDM in older men and women, independent of glucose intolerance, family history of diabetes, hypertension medication use, and waist-to-hip ratio but not independent of fasting triglyceride or insulin levels. Small, dense LDL was highly correlated with increased plasma triglyceride and decreased HDL cholesterol, and a composite variable of these three lipoproteins identified by principal-components analysis was also associated with the development of NIDDM. These results demonstrate that small, dense LDL is a risk factor for NIDDM and may contribute to risk of atherosclerosis among prediabetics.


*    Selected Abbreviations and Acronyms
 
BMI = body mass index
CHD = coronary heart disease
DAMPN = 3-dimethylaminopropionitrile
NIDDM = non–insulin dependent diabetes mellitus
OGTT = oral glucose tolerance test
OR = odds ratio
PPD = peak particle diameter
WHO = World Health Organization


*    Acknowledgments
 
This study was supported by a grant from the Academy of Finland and was performed during Dr Austin's tenure as an Established Investigator of the American Heart Association. The authors would like to thank Jon Diemer, Andy Louie, Hal Kennedy, and Anitra Wolf for excellent technical support and Drew Levy for assistance with data analysis.

Received March 20, 1995; revision received April 24, 1995; accepted April 25, 1995.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Pyörälä K, Laakso M, Uusitupa M. Diabetes and atherosclerosis: an epidemiological view. Diabetes Metab Rev. 1987;3:463-524. [Medline] [Order article via Infotrieve]
  2. Barrett-Connor E, Orchard T. Diabetes and heart disease. In: National Diabetes Data Group. Diabetes in America. Bethesda, Md: National Institutes of Health; 1985; XVI:1-41. NIH publication 85-1468.
  3. Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK. Cardiovascular risk factors in confirmed prediabetic individuals: does the clock for coronary heart disease start ticking before the onset of clinical diabetes? JAMA. 1990;263:2893-2898. [Abstract]
  4. McPhillips JB, Barrett-Connor E, Wingard DL. Cardiovascular disease risk factors prior to the diagnosis of impaired glucose tolerance and non-insulin dependent diabetes mellitus in a community of older adults. Am J Epidemiol. 1990;131:443-453. [Abstract/Free Full Text]
  5. Mykkänen L, Kuusisto J, Pyörälä K, Laakso M. Cardiovascular disease risk factors as predictors of type 2 (non-insulin dependent) diabetes mellitus in elderly subjects. Diabetologia. 1993;36:553-559. [Medline] [Order article via Infotrieve]
  6. Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WS, Krauss RM. Low-density lipoprotein subclass patterns and risk of myocardial infarction. JAMA. 1988;260:1917-1921. [Abstract]
  7. Austin MA, Hokanson JE, Brunzell JD. Characterization of low-density lipoprotein subclasses: methodologic approaches and clinical relevance. Curr Opin Lipidol. 1994;5:395-405. [Medline] [Order article via Infotrieve]
  8. Reaven GM. Banting Lecture 1988: Role of insulin resistance in human disease. Diabetes. 1988;37:1595-1607. [Abstract]
  9. Stout RW. Insulin and atheroma: 20-yr perspective. Diabetes Care. 1990;13:631-654. [Abstract]
  10. Laakso M. Dyslipidemia, insulin resistance and atherosclerosis. Ann Med. 1992;24:505-509. [Medline] [Order article via Infotrieve]
  11. Stern MP. The insulin resistance syndrome. In: Alberti KGMM, DeFronzo RA, Zimmett P, eds. International Textbook of Diabetes Mellitus. Chichester, UK: Wiley. In press.
  12. Selby JV, Austin MA, Newman B, Zhang D, Quesenberry CP Jr, Mayer EJ, Krauss RM. LDL subclass phenotypes and the insulin resistance syndrome. Circulation. 1993;88:318-387.
  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. 1993;92:141-146.
  14. Haffner SM, Mykkänen L, Paidi M, Valdez R, Howard BV, Stern MP. Small dense LDL is associated with the insulin resistance syndrome in non-diabetic subjects. Diabetologia. In press.
  15. Austin MA, King M-C, Vranizan KM, Krauss RM. Atherogenic lipoprotein phenotype: a proposed genetic marker for coronary heart disease risk. Circulation. 1990;82:495-506. [Abstract/Free Full Text]
  16. Swinkels DWL, Demacker PNM, Hendriks JCM, van 't Laar A. Low density lipoprotein subfractions and relationship to other risk factors for coronary artery disease in healthy individuals. Arteriosclerosis. 1989;9:604-613. [Abstract/Free Full Text]
  17. McNamara JR, Jenner JL, Li Z, Wilson PWF, Schaefer EJ. Change in LDL particle size is associated with change in plasma triglyceride concentration. Arterioscler Thromb. 1992;12:1284-1290. [Abstract/Free Full Text]
  18. Haffner SM, Mykkanen L, Valdez RA, Paidi M, Stern MP, Howard BV. LDL size and subclass pattern in a biethnic population. Arterioscler Thromb. 1993;13:1623-1630. [Abstract/Free Full Text]
  19. Barakat HA, Carpenter JW, McLendon VD, Khazanie P, Leggett N, Heath J, Marks R. Influence of obesity, impaired glucose tolerance, and NIDDM on LDL structure and composition: possible link between hyperinsulinemia and atherosclerosis. Diabetes. 1990;39:1527-1533. [Abstract]
  20. Peeples LH, Carpenter JW, Israel RG, Barakat HA. Alterations in low-density lipoproteins in subjects with abdominal adiposity. Metabolism. 1989;38:1029-1036. [Medline] [Order article via Infotrieve]
  21. Terry RB, Wood PD, Haskell WL, Stafanick ML, Krauss RM. Regional adiposity patterns in relation to lipids, lipoprotein cholesterol, and lipoprotein subfraction mass in men. J Clin Endocrinol Metab. 1989;68:191-199. [Abstract]
  22. Fujimoto WF, Abbate SL, Kahn SE, Hokanson JE, Brunzell JD. The visceral adiposity syndrome in Japanese-American men. Obesity Res. 1994;2:364-371. [Medline] [Order article via Infotrieve]
  23. Feingold KR, Grunfeld C, Pang M, Coerrler W, Krauss RM. LDL subclass phenotypes and triglyceride in non-insulin-dependent diabetes. Arterioscler Thromb. 1992;12:1496-1502. [Abstract/Free Full Text]
  24. Stewart MS, Laker MF, Dyer RG, Game F, Mitcheson J, Winocour PH, Alberti KGMM. Lipoprotein compositional abnormalities and insulin resistance in type II diabetic patients with mild hyperlipidemia. Arterioscler Thromb. 1993;13:1046-1051. [Abstract/Free Full Text]
  25. Mykkänen L, Laakso M, Uusitupa M, Pyörälä K. Prevalence of diabetes and impaired glucose tolerance in elderly subjects and their association with obesity and family history of diabetes. Diabetes Care. 1990;13:1099-1105. [Abstract]
  26. Mykkänen L, Laakso M, Penttilä I, Pyörälä K. Asymptomatic hyperglycemia and cardiovascular risk factors in the elderly. Atherosclerosis. 1991;88:153-161. [Medline] [Order article via Infotrieve]
  27. World Health Organization. Diabetes mellitus: report of a WHO Study Group. Technical Report Series No. 727. Geneva, Switzerland: World Health Organization; 1985.
  28. Krauss RM, Burke DJ. Identification of multiple subclasses of plasma low-density lipoproteins in normal humans. J Lipid Res. 1982;23:97-104. [Abstract]
  29. Nichols AV, Krauss RM, Musliner TA. Nondenaturing polyacrylamide gradient gel electrophoresis. In: Segrest JP, Albers JJ, eds. Methods in Enzymology, Volume 128, Plasma Lipoproteins, Part A. Preparation, Structure and Molecular Biology. Orlando, Fla: Academic Press; 1986:417-431.
  30. Rainwater DL, Andres DW, Ford AL, Lowe WF, Blanche PJ, Krauss RM. Production of polyacrylamide gradient gels for the electrophoretic resolution of lipoproteins. J Lipid Res. 1992;33:1876-1881. [Abstract]
  31. Austin MA, King M-C, Vranizan KM, Newman B, Krauss RM. Inheritance of low density lipoprotein subclass patterns: results of complex segregation analysis. Am J Hum Genet. 1988;43:838-846. [Medline] [Order article via Infotrieve]
  32. Winer BJ. Statistical Principles in Experimental Design. New York, NY: McGraw-Hill Book Co; 1971.
  33. Berenson ML. Intermediate Statistical Methods and Applications. Inglewood, NJ: Prentice-Hall Inc; 1983:102-104.
  34. Schlesselman JJ. Case Control Studies: Design, Conduct, Analysis. New York, NY: Oxford University Press Inc; 1982.
  35. Edwards KL, Austin MA, Newman B, Mayer EJ, Krauss RM, Selby JV. Multivariate analysis of the insulin resistance syndrome. Arterioscler Thromb. 1994;14:1940-1945. [Abstract/Free Full Text]
  36. Stevens J. Applied Multivariate Statistics for the Social Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates; 1986.
  37. Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and Other Multivariate Methods. Boston, Mass: Kent Publishing Co; 1988.
  38. SAS User's Guide: Version 5 Edition. Cary, NC: SAS Institute Inc; 1985.
  39. Galeano NF, Milne R, Marcel YL, Walsh MT, Levy E, Nguyen TD, Gleeson A, Arad Y, Witte L, Alhaideri M, Runsey SC, Deckelbaum RJ. Apolipoprotein-B structure and receptor recognition of triglyceride-rich low-density lipoprotein (LDL) is modified in small LDL but not in triglyceride-rich LDL of normal size. J Biol Chem. 1994;269:511-519. [Abstract/Free Full Text]
  40. Ferrannini E, Haffner SM, Mitchell BD, Stern MP. Hyperinsulinaemia: the key feature of a cardiovascular and metabolic syndrome. Diabetologia. 1991;34:416-422.[Medline] [Order article via Infotrieve]
  41. Howard BV. Lipoprotein metabolism in diabetes. Curr Opin Lipidol. 1994;5:216-220. [Medline] [Order article via Infotrieve]
  42. Lahdenperä S, Tilly-Kiesi M, Vuorinen-Markkola H, Kuusi T, Taskinen M-R. Effects of gemfibrozil on low-density lipoprotein particle size, density distribution, and composition in patients with type II diabetes. Diabetes Care. 1993;16:584-592. [Abstract]
  43. West KM, Ahuja MMS, Bennett PH, Czyzyk A, De Acosta OM, Fuller JH, Grab B, Grabauskas V, Jarrett RJ, Kosaka K, Keen H, Krowlewski AS, Miki E, Schliack V, Teuscher A, Watkins PJ, Stober JA. The role of circulating glucose and triglyceride concentrations and their interactions with other `risk factors' as determinants of arterial disease in nine diabetic population samples from the WHO Multinational Study. Diabetes Care. 1983;6:361-368. [Abstract]
  44. Bergstrom RW, Leonetti DL, Newell-Morris LL, Shuman WP, Wahl PW, Fujimoto WY. Association of plasma triglyceride and C-peptide with coronary heart disease in Japanese-American men with a high prevalence of glucose intolerance. Diabetologia. 1990;33:489-496. [Medline] [Order article via Infotrieve]
  45. Fontbonne A, Eschwege E, Cambien F, Richard J-L, Ducimetiere, Thibult N, Warnet J-M, Claude J-R, Rosselin G-E. Hypertriglyceridemia as a risk factor of coronary heart disease mortality in subjects with impaired glucose tolerance or diabetes: results from the 11-year follow-up of the Paris Prospective Study. Diabetologia. 1989;2:300-304.
  46. Laws A, Marcus EB, Grove JS, Curb JD. Lipid and lipoproteins as risk factors for coronary heart disease in men with abnormal glucose tolerance: the Honolulu Heart Program. J Intern Med. 1993;234:471-478. [Medline] [Order article via Infotrieve]
  47. Laakso M, Lehto S, Penttilä I, Pyörälä K. Lipids and lipoproteins predicting coronary heart disease mortality and morbidity in patients with non-insulin-dependent diabetes. Circulation. 1993;88:1421-1430. [Abstract/Free Full Text]
  48. Fisher WR. Heterogeneity of plasma low density lipoproteins: manifestations of the physiologic phenomenon in man. Metabolism. 1983;32:283-291. [Medline] [Order article via Infotrieve]
  49. Crouse JR, Parks JS, Schey HM. Studies of low density lipoprotein molecular weight in human beings with coronary artery disease. J Lipid Res. 1985;26:566-574. [Abstract]
  50. Griffin BA, Caslake MJ, Yip B, Tait GW, Packard CJ, Shepherd J. Rapid isolation of low density lipoprotein (LDL) subfractions from plasma by density gradient ultracentrifugation. Atherosclerosis. 1990;83:59-67. [Medline] [Order article via Infotrieve]
  51. Tornvall P, Karpe F, Carlson LA, Hamsten A. Relationship of low density lipoprotein subfractions to angiographically defined coronary artery disease in young survivors of myocardial infarction. Atherosclerosis. 1991;90:67-80. [Medline] [Order article via Infotrieve]
  52. Campos H, Genest JJ, Blijlevens E, McNamara JR, Jenner JL, Ordovas JM, Wilson PWF, Schaefer EJ. Low density lipoprotein particle size and coronary artery disease. Arterioscler Thromb. 1992;12:187-195. [Abstract/Free Full Text]
  53. Jaakkola O, Solakivi T, Tertov VV, Orekhov AN, Miettinen TA, Nikkari T. Characteristics of low-density lipoprotein subfractions from patients with coronary artery disease. Coron Artery Dis. 1993;4:379-385. [Medline] [Order article via Infotrieve]
  54. Coresh J, Kwiterovich PO, Bachorik P. Association of low density lipoprotein size, density and composition with premature coronary artery disease. J Lipid Res. 1993;34:1687-1697. [Abstract]
  55. Griffin BA, Freeman DJ, Tait GW, Thomson J, Caslake MJ, Packard CJ, Shepherd J. Role of plasma triglyceride in the regulation of plasma low density lipoprotein (LDL) subfractions: relative contribution of small, dense LDL to coronary heart disease risk. Atherosclerosis. 1994;106:241-253. [Medline] [Order article via Infotrieve]
  56. Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK, Ferrannini E. Parental history of diabetes is associated with increased cardiovascular risk factors. Arteriosclerosis. 1989;9:928-933. [Abstract/Free Full Text]
  57. Sarlund H, Pyorala K, Penttila I, Laakso M. Early abnormalities in coronary heart disease risk factors in relatives of subjects with non-insulin-dependent diabetes. Arterioscler Thromb. 1992;12:657-663. [Abstract/Free Full Text]
  58. Austin MA. Genetics of low-density lipoprotein subclasses. Curr Opin Lipidol. 1993;4:125-132.
  59. Austin MA, Newman B, Selby JV, Edwards K, Mayer E, Krauss RM. Genetics of low-density lipoprotein subclass phenotypes in women twins: concordance, heritability, and commingling analysis. Arterioscler Thromb. 1993;13:687-695. [Abstract/Free Full Text]
  60. Nishina PM, Johnson JP, Naggert KJ, Krauss RM. Linkage of atherogenic lipoprotein phenotype to the low density lipoprotein receptor locus on the short arm of chromosome 19. Proc Natl Acad Sci U S A. 1992;89:708-712.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
I. Aeberli, M. B Zimmermann, L. Molinari, R. Lehmann, D. l'Allemand, G. A Spinas, and K. Berneis
Fructose intake is a predictor of LDL particle size in overweight schoolchildren
Am. J. Clinical Nutrition, October 1, 2007; 86(4): 1174 - 1178.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart J SupplHome page
A. S. Wierzbicki
Diabetic dyslipidaemia: the triad
Eur. Heart J. Suppl., October 1, 2006; 8(suppl_F): F30 - F33.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
A. Festa, K. Williams, A. J.G. Hanley, J. D. Otvos, D. C. Goff, L. E. Wagenknecht, and S. M. Haffner
Nuclear Magnetic Resonance Lipoprotein Abnormalities in Prediabetic Subjects in the Insulin Resistance Atherosclerosis Study
Circulation, June 28, 2005; 111(25): 3465 - 3472.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
M. D. Badzioch, R. P. Igo Jr, F. Gagnon, J. D. Brunzell, R. M. Krauss, A. G. Motulsky, E. M. Wijsman, and G. P. Jarvik
Low-Density Lipoprotein Particle Size Loci in Familial Combined Hyperlipidemia: Evidence for Multiple Loci From a Genome Scan
Arterioscler. Thromb. Vasc. Biol., October 1, 2004; 24(10): 1942 - 1950.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
B. Todorova, A. Kubaszek, J. Pihlajamaki, J. Lindstrom, J. Eriksson, T. T. Valle, H. Hamalainen, P. Ilanne-Parikka, S. Keinanen-Kiukaanniemi, J. Tuomilehto, et al.
The G-250A Promoter Polymorphism of the Hepatic Lipase Gene Predicts the Conversion from Impaired Glucose Tolerance to Type 2 Diabetes Mellitus: The Finnish Diabetes Prevention Study
J. Clin. Endocrinol. Metab., May 1, 2004; 89(5): 2019 - 2023.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
M. A. Austin, K. L. Edwards, M. J. McNeely, W. L. Chandler, D. L. Leonetti, P. J. Talmud, S. E. Humphries, and W. Y. Fujimoto
Heritability of Multivariate Factors of the Metabolic Syndrome in Nondiabetic Japanese Americans
Diabetes, April 1, 2004; 53(4): 1166 - 1169.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
M. A. Austin, K. L. Edwards, S. A. Monks, K. M. Koprowicz, J. D. Brunzell, A. G. Motulsky, M. C. Mahaney, and J. E. Hixson
Genome-wide scan for quantitative trait loci influencing LDL size and plasma triglyceride in familial hypertriglyceridemia
J. Lipid Res., November 1, 2003; 44(11): 2161 - 2168.
[Abstract] [Full Text] [PDF]

Home page
 Diabetes CareHome page
K. Winkler, T. Konrad, S. Fullert, I. Friedrich, R. Destani, M. W. Baumstark, K. Krebs, H. Wieland, and W. Marz
Pioglitazone Reduces Atherogenic Dense LDL Particles in Nondiabetic Patients With Arterial Hypertension: A double-blind, placebo-controlled study
Diabetes Care, September 1, 2003; 26(9): 2588 - 2594.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
K. K. Berneis and R. M. Krauss
Metabolic origins and clinical significance of LDL heterogeneity
J. Lipid Res., September 1, 2002; 43(9): 1363 - 1379.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
C.F. Ebenbichler, M. Laimer, S. Kaser, A. Ritsch, A. Sandhofer, H. Weiss, F. Aigner, and J.R. Patsch
Relationship Between Cholesteryl Ester Transfer Protein and Atherogenic Lipoprotein Profile in Morbidly Obese Women
Arterioscler. Thromb. Vasc. Biol., September 1, 2002; 22(9): 1465 - 1469.
[Abstract] [Full Text] [PDF]

Home page
 Diabetes CareHome page
T. Kazumi, A. Kawaguchi, K. Sakai, T. Hirano, and G. Yoshino
Young Men With High-Normal Blood Pressure Have Lower Serum Adiponectin, Smaller LDL Size, and Higher Elevated Heart Rate Than Those With Optimal Blood Pressure
Diabetes Care, June 1, 2002; 25(6): 971 - 976.
[Abstract] [Full Text] [PDF]