Ratio of remnant-like particle-cholesterol to serum total triglycerides is an effective alternative to ultracentrifugal and electrophoretic methods in the diagnosis of familial type III hyperlipoproteinemia.
apo E-2/2, an uncommon apoE phenotype seen in ~1% of the general population, has reduced affinity for receptors that are necessary for removal of remnant lipoproteins (3). When combined with other genetic or environmental factors, apoE-2 homozygotes exhibit hyperlipidemia with substantial accumulation of [beta]-VLDL. In these subjects, both genetic and nongenetic factors contribute to the development of overt hyperlipidemia with substantial increases of total cholesterol (TC) and triglycerides (TGs) in plasma. Compared with subjects with other types of HLP, subjects with type III HLP usually are very responsive to therapy. In most cases, life-style changes alone or in combination with treatment of a preexisting metabolic condition will normalize plasma lipid values (3). Early diagnosis of this familial disorder will therefore help physicians to select the most appropriate therapy and to identify other affected family members. It is also beneficial for these subjects to modify their life-styles to prevent premature onset of cardiovascular disease.
No single simple diagnostic test for type III HLP is available. Electrophoretic demonstration of [beta]-VLDL in total VLDL (lipoproteins with a hydrated density, d, <1.006 kg/L) is considered the gold standard for diagnosing this disorder (1-5). A VLDL-cholesterol (VLDL-C)-to-serum total TG molar ratio of [greater than or equal to] 0.69 ([greater than or equal to] 0.30 when using mg/dL) reflects the presence of cholesterol-rich [beta]-VLDL and is also considered diagnostic of type III HLP (3,5). apoE-2 homozygosity can confirm the diagnosis. However, only a small fraction of apoE-2 homozygotes are hyperlipidemic and, rarely, other mutant apoE alleles can cause dominant forms of type III HLP (3).
To detect [beta]-VLDL and compute the VLDL-C-to-total TG ratio, UC must be used. UC is complex, labor-intensive, and not readily available in routine clinical laboratories. More convenient alternatives to UC and AGE are therefore needed. One convenient candidate alternative is the RLP-Cholesterol Immunoseparation Assay (RLP-Cholesterol Assay). This assay isolates remnant-like particles (RLPs) from human serum or plasma, using an immunoaffinity gel containing monoclonal antibodies to human apoA-I and apoB-100 (6, 7). These monoclonal antibodies are conjugated to Sepharose-4B beads, facilitating the separation of RLPs from other lipoproteins (HDL, LDL, nascent VLDL, and others). Cholesterol in the unbound fraction [remnant-like particle-cholesterol (RLP-C)] is then quantified by an enzymatic spectrophotometric method. It has been demonstrated that remnant lipoproteins isolated by this method resemble [beta]-VLDL (6, 8, 9). They are enriched in cholesteryl esters and apoE and have reduced mobility on AGE (6,8-11); they are also rapidly taken up by macrophages without modification (12). RLP-C concentrations are significantly higher in patients with type III HLP than subjects without this disorder (6, 8, 9,11,13). Data from these studies also suggest that the RLP-C-to-total TG ratio is also higher in subjects with type III HLP. The purpose of the current study was to compare the ability of the RLP-Cholesterol Assay with UC and AGE for the diagnosis of type III HLP in a group of hypertriglyceridemic subjects.
Materials and Methods
STUDY SUBJECTS AND DESIGN
To establish a reference interval for the RLP-C-to-total TG ratio, data from a previous study (14), 364 men and 362 women 18 years of age and older were reanalyzed. These subjects were either normolipidemic or hyperlipidemic but were free of symptoms and signs of cardiovascular disease and endocrine or metabolic disorders that may affect lipid metabolism.
To compare the ability of the RLP-Cholesterol Assay with other current methods to identify patients with type III HLP, hypertriglyceridemic subjects with and without type III HLP were enrolled from two lipid clinics specialized in the treatment of familial HLP. These two clinics were at the Clinical Research Institute of Montreal (Montreal, Quebec, Canada) and at the Department of Cardiovascular Genetics Research, University of Utah (Salt Lake City, UT). All eligible subjects were 18 years and older and had a fasting serum total TG concentration between 2.26 and 9.04 mmol/L .  They took no lipid-lowering medication for at least 6 weeks before sample collection. All subjects with type III HLP had [beta]-VLDL, whereas all other subjects lacked this marker. The LDL-cholesterol (LDL-C) concentration was >4.16 mmol/L in subjects classified as type IIb HLP and <4.16 mmol/L in subjects classified as type IV HLP. This study used a case-control design. Subjects with type III HLP were enrolled first. For each subject with type III HLP enrolled, one subject with type IIb HLP and one subject with type IV HLP of the same gender and similar age ([+ or -] 5 years) were also enrolled. Study subjects were not enrolled on the basis of their historical serum TC value. However, subjects with type IIb or type IV HLP were selected to match the subjects with type III HLP as closely as possible in their fasting serum TG concentrations. Serum samples were collected from subjects fasted overnight and sent to Pacific Biometrics, Inc. (Seattle) for analysis. Final classification of the study subjects was based on the detection of [beta]-VLDL by AGE. apoE genotyping (for samples from Salt Lake City) or phenotyping (for samples from Montreal) was carried out on all samples.
Data were also reanalyzed from 749 subjects enrolled in a previously reported study carried out in 14 medical centers in the US and Canada (14). Approximately one-third of the subjects had angiographically demonstrated coronary artery disease and two-thirds were healthy controls free of coronary artery disease. Because serum lipid values were not part of the inclusion/ exclusion criteria, these subjects were either normolipidemic or hyperlipidemic. Only the hyperlipidemic subgroup of the study subjects (fasting TC concentrations >5.18 mmol/L and TG concentrations >2.26 mmol/L; n = 140) was included in this secondary analysis.
All clinical study protocols and informed consent forms were approved by the local Institutional Review Boards. Informed consent forms were signed by all participating subjects.
LIPID AND LIPOPROTEIN ANALYSIS
Serum TC and TGs were measured by CDC-standardized enzymatic methods. HDL-cholesterol (HDL-C) was measured after precipitation of non-HDL lipoproteins by dextran sulfate (MT 50 000)-[Ca.sup.2+]. The RLP-Cholesterol Assay kits were provided by Japan Immunoresearch Laboratories Co., Ltd. The performance of this assay had been evaluated previously (13-15).
VLDL (d <1.006 kg/L) and non-VLDL lipoproteins (d [greater than or equal to] 1.006 kg/L) were separated from serum samples by UC at 100 000g by a Beckman Ti 42.2 rotor for 4 h as described previously (14,16). VLDL, non-VLDL, and unfractionated serum were then analyzed by the REP [TM] Cholesterol Profile-15 Kit (a commercial AGE device manufactured by Helena Laboratories). [beta]-VLDL, which has density <1.006 kg/L (like VLDL) and [beta] mobility on AGE (like LDL), can be detected by this kit. The electrophoresis pattern of each sample was interpreted by three analysts to maintain consistency. In addition, [beta]-VLDL-cholesterol ([beta]-VLDL-C) was also estimated as the VLDL-C-to-total TG ratio, and quantified by using the following formula (17):
Estimated [beta]-VLDL-C (mg/dL) = VLDL-C - 0.17(total TG) / 0.521
A VLDL-C-to-total TG ratio [greater than or equal to] 0.69 (3,4) and an estimated [beta]-VLDL-C concentration >1.04 mmol/L (17) suggest the presence of [beta]-VLDL and therefore type III HLP.
apoE phenotyping was performed by isoelectric focusing gel electrophoresis followed by immunoblotting of whole serum (18). apoE genotyping was performed by a restriction isotyping method (19).
Differences of lipid analytes between the type III group and the non-type III group were evaluated statistically by the Mann-Whitney U-test because most analytes, particularly TRLs and their remnants, have nongaussian distributions. The correlation between different analytes was assessed by the Pearson correlation coefficients. Linear relation between two analytes was evaluated by a linear regression model. Differences between regression lines were evaluated by analysis of variance. The diagnostic sensitivity, specificity, and accuracy of each test were calculated using standard formulas (20).
REFERENCE INTERVAL FOR THE RLP-C-TO-TOTAL TG RATIO
In 364 healthy men and 362 healthy women described previously (14), the median fasting TC, TG, LDL-cholesterol (LDL-C), HDL-C, and RLP-C concentrations were 5.21, 1.10, 3.34, 1.22, and 0.13 mmol/L, respectively. The median and 95th percentile RLP-C-to-total TG ratios were 0.11 and 0.18, respectively, for both fasting and random samples. A 95th percentile of 0.22, however, was observed in several subgroups, which accounted for a rather large portion of the study subjects.
GENERAL CHARACTERISTICS OF STUDY SUBJECTS
Seventy-one hypertriglyceridemic subjects, all Caucasians, were enrolled either in Montreal or from Salt Lake City. Sixty-two subjects (23 type III, 20 type IIb, and 19 type IV) with fasting total TG concentrations between 2.26 and 9.04 mmol/L met the inclusion criteria of the study protocol. The general characteristics of these 62 subjects are shown in Table 1. The remaining nine subjects had gross hypertriglyceridemia (TG concentration >9.04 mmol/L) and did not meet the inclusion criteria of the study protocol. Samples from these nine subjects were analyzed separately to evaluate the applicability of each method in subjects with severe hypertriglyceridemia.
LIPID PROFILE OF STUDY SUBJECTS
As shown in Table 2, serum TC and HDL-C concentrations of the subjects with type III HLP and subjects with IIb HLP were similar. The serum LDL-C and HDL-C concentrations of the subjects with type III HLP and subjects with type IV HLP were also similar. Despite efforts to match serum total TG concentrations of subjects without type III HLP with those with type III HLP, the median total TG concentration was significantly higher in those with type III HLP. Subjects with type III HLP had significantly higher VLDL-C, estimated [beta]-VLDL-C, and RLP-C concentrations than those without type III HLP.
RATIOS OF VLDL-C AND RLP-C TO TOTAL TG
At any given total TG concentration between 2.26 and 9.04 mmol/L, VLDL-C and RLP-C concentrations were higher in subjects with type III HLP than those with type IIb or IV HLP (Fig. 1). The linear relation between VLDL-C and total TGs was VLDL-C = 0.89(TG) - 0.42 for subjects with type III HLP and VLDL-C = 0.34(TG) + 0.18 for the others. The slopes of the two linear regressions were different (P <0.01). In comparison, the linear relation between RLP-C and total TGs was RLP-C = 0.37(TG) - 0.02 for subjects with type III HLP and RLP-C = 0.14(TG) + 0.02 for the others. The slopes of these two linear regressions also differed significantly (P <0.05). The mean fasting VLDL-C-to-total TG ratios were 0.80 [+ or -] 0.22 for subjects with type III HLP and 0.39 [+ or -] 0.10 for the others (P <0.001); the RLP-C to total TG ratios were 0.36 [+ or -] 0.14 for subjects with type III HLP and 0.15 [+ or -] 0.05 for the others (P <0.001). Plots of individual ratios of VLDL-C and RLP-C to total TGs are shown in Fig. 2.
[FIGURE 1 OMITTED]
PERFORMANCE OF THE RLP-C ASSAY AND OTHER METHODS IN DIAGNOSING TYPE III HLP
On the basis of their fasting lipid profile, the 71 enrolled hypertriglyceridemic subjects could be segregated into one of three groups: subjects (3 type III and 8 non-type III) with isolated hypertriglyceridemia (TC concentrations <5.18 mmol/L), subjects (20 type III and 31 non-type III) with hypercholesterolemia and with mild to moderate hypertriglyceridemia (TC concentrations >5.18 mmol/L and TG concentrations between 2.26 and 9.04 mmol/L), and subjects (5 type III and 4 non-type III) with gross hypertriglyceridemia (TG concentrations >9.04 mmol/L).
[FIGURE 2 OMITTED]
The optimal performance of the RLP-C-to-total TG ratio was achieved in subjects with hypercholesterolemia and with mild to moderate hypertriglyceridemia. Fifty-one subjects (20 with type III HLP and 31 without type III HLP) were in this group. All 20 subjects with type III HLP were apoE-2 homozygotes and had detectable [beta]-VLDL and estimated [beta]-VLDL-C concentrations >1.04 mmol/L (Table 3). Nineteen and 16 of these subjects (95.0% and 80.0%, respectively) also had a RLP-C-to-total TG ratio [greater than or equal to] 0.23 and a VLDL-C-to-total TG ratio [greater than or equal to] 0.69, respectively. Among the 31 subjects without type III HLP, none had detectable [beta]-VLDL. Thirty (96.8%) of these subjects also had a VLDL-C-to-total TG ratio <0.69. Twenty-nine (93.5%) had estimated [beta]-VLDL-C concentrations [less than or equal to] 1.04 mmol/L and a RLP-C-to-total TG ratio <0.23. Compared with the gold standard (AGE of VLDL), estimated [beta]-VLDL, the RLP-C-to-total TG ratio, and the VLDL-C-to-total TG ratio correctly classified 49 (96.1%), 48 (94.1%), and 46 (90.2%) of the 51 subjects (Table 3).
Among the 11 subjects with isolated hypertriglyceridemia, none of the 3 with type III HLP had a VLDL-C to total TG ratio [greater than or equal to] 0.69. Only one had a RLP-C-to-total TG ratio [greater than or equal to] 0.23 and an estimated [beta]-VLDL concentration >1.06 mmol/L. In comparison, none of the eight subjects without type III HLP had an estimated [beta]-VLDL-C concentration and a VLDL-C-to-total TG ratio exceeding the respective cutoff value. Seven also had a RLP-C-to-total TG ratio <0.23.
Among the nine subjects with hypercholesterolemia and gross hypertriglyceridemia, all five with type III HLP had estimated [beta]-VLDL-C concentrations, VLDL-C-to-total TG ratios, and RLP-C-to-total TG ratios greater than the respective cutoff values. None of the four subjects without type III HLP had an estimated [beta]-VLDL-C concentration or VLDL-C-to-total TG ratio greater than the respective cutoff values. Two of these subjects, however, had a RLP-C-to-total TG ratio [greater than or equal to] 0.23.
Similar findings were observed in 140 hyperlipidemic subjects from a previously reported study (14). Two of the 140 hyperlipidemic subjects had type III HLP evidenced by the presence of the apoE-2/2 phenotype. In both, the RLP-C-to-total TG ratio and the estimated [beta]-VLDL-C concentration exceeded their respective cutoff values. Only one had a VLDL-C-to-total TG ratio [greater than or equal to] 0.69. Three samples from two of the subjects without type III HLP (one fasting and two random) had TG concentrations >9.04 mmol/L. All three samples had RLP-C-to-total TG ratios [greater than or equal to] 0.23. Two had estimated [beta]-VLDL-C concentrations exceeding 1.04 mmol/L. One of the two also had a VLDL-C-to-total TG ratio [greater than or equal to] 0.69. Among the 135 remaining hyperlipidemic subjects without type III HLP, the specificity of VLDL-C-to-total TG ratio, the RLP-C-to-total TG ratio, and the estimated [beta]-VLDL-C concentration was 100%, 99%, and 96%, respectively.
We have compared a new method utilizing the RLP-Cholesterol Assay with existing methods for the diagnosis of type III HLP in 28 patients with type III HLP and 43 hypertriglyceridemic subjects without this disorder. In this case-control study, subjects with type IIb or type IV HLP were selected as controls for subjects with type III HLP because these subjects generally have similar lipid profiles (TC, TGs, and HDL-C). Without sophisticated analytical methods such as UC (for VLDL isolation) and AGE (for [beta]-VLDL measurement), type III HLP cannot be distinguished reliably from two other types of HLP by laboratory criteria. This study showed that (a) VLDL-C and RLP-C concentrations were significantly higher in subjects with type III HLP than in subjects without this disorder; and (b) the RLP-C-to-total TG ratio had an ability similar to [beta]-VLDL (measured by AGE or estimated) and was superior to the VLDL-C-to-total TG ratio in diagnosing type III HLP in a group of subjects with hypercholesterolemia and mild to moderate hypertriglyceridemia.
Although VLDL-C concentrations were significantly higher in subjects with type III HLP than in hypertriglyceridemic subjects without this disorder, there was appreciable overlap between the two groups. For this reason, the VLDL-C-to-total TG ratio, instead of VLDL-C per se, has been recommended in current practice to diagnose type III HLP (3-5). The same is true for RLP-C. Whereas VLDL-C and RLP-C concentrations alone provide information on the magnitude of the increase in remnant lipoproteins in plasma, the VLDL-C-to-total TG and the RLP-C-to-total TG ratios provide an estimate of the enrichment of TRL in cholesterol, a characteristic of [beta]-VLDL. This study has shown that, although RLP-C and VLDL-C were highly correlated, the RLP-C-to-total TG ratio has slightly better ability, particularly in its sensitivity, than the currently accepted marker (VLDL-C-to-total TG ratio) in diagnosing type III HLP. The "VLDL" fraction separated by UC is actually TRLs with hydrated densities <1.006 kg/L and includes chylomicrons, chylomicron remnants, VLDL, and VLDL remnants. Earlier research has shown that, compared with lipoproteins bound to monoclonal antibody JI-H, lipoproteins not recognized by this monoclonal antibody have certain physical and chemical properties of remnant lipoproteins [e.g., enrichment in cholesteryl esters and apoE, and reduced mobility on AGE (6, 8-11)]. Better utility of the RLP-C-to-total TG ratio than the VLDL-C-to-total TG ratio may reflect the remnant-like properties of TRLs that are included in the RLP-C measurement.
The gold standard for diagnosing type III HLP is the detection of [beta]-VLDL by AGE. To use this standard, two consecutive steps must be taken: UC of serum lipoproteins, followed by AGE of two lipoprotein fractions (d <1.006 and d [greater than or equal to] 1.006 kg/L) together with unfractionated serum or plasma. This adds to the cost and complexity to the already inconvenient UC method. Furthermore, interpretation of [beta]-VLDL by AGE requires an experienced individual. Even then, the interpretation can be subjective. In comparison, isolation of RLPs requires only a small mixing device (7) and does not require a high degree of technical skill. Any clinical laboratory with an accurate and precise automated, open-system design chemistry analyzer can perform the RLP-C analysis in a relatively short period. The ability to use previously frozen samples (14) also offers a substantial advantage over detection of [beta]-VLDL. Without the use of either AGE or UC, the RLP-C-to-total TG ratio achieved 94% diagnostic accuracy in this case-control study (Table 3). In a group of 140 hyperlipidemic subjects nonselective for type III HLP, the RLP-C-to-total TG ratio correctly classified 138 of them.
Neither the VLDL-C-to-total TG ratio nor the RLP-C-to-total TG ratio should be used to diagnose type III HLP in a general population. The VLDL-C-to-total TG ratio has been recognized as unreliable in subjects without hyperlipidemia (3, 4) and in subjects with serum total TG concentrations exceeding 11.3 mmol/L (21). As with the VLDL-C-to-total TG ratio, the RLP-C-to-total TG ratio is more useful in patients with hypercholesterolemia and with mild to moderate hypertriglyceridemia, a potential indicator of substantial accumulation of [beta]-VLDL (Table 3 and Fig. 2). More false-negative results occurred with calculated values (estimated [beta]-VLDL-C, VLDL-C-to-total TG ratio, and RLP-C-to-total TG ratio) in patients without hypercholesterolemia. The accumulation of [beta]-VLDL in these patients is likely insufficient to permit detection by the calculated methods. On the other hand, the RLP-C-to-total TG ratio had more false-positive results in patients with gross hypertriglyceridemia. Subjects with gross hypertriglyceridemia may have type III, type I, or type V HLP. The biochemical features of type I and type V HLP are gross increases in TG concentrations (typically >9.04 mmol/L) and accumulation of apoB-48-containing chylomicrons and chylomicron remnants as well as apoB-100containing VLDL. Use of a monoclonal antibody that recognizes an epitope within the apoB-51 region distal to the C-terminus of apoB-48 (7) allows apo B-48-containing lipoproteins to be selectively enriched in the RLP fraction compared with the VLDL fraction isolated by the UC. This may yield RLP-C-to-total TG ratios [greater than or equal to] 0.23.
In conclusion, the RLP-C-to-total TG ratio provides a convenient and effective alternative to methods that require UC for diagnosing type III HLP. Optimal performance of the RLP-C-to-total TG ratio can be achieved in target populations consisting of subjects with hypercholesterolemia (TC concentration >5.18 mmol/L) and with mild to moderate hypertriglyceridemia (TG concentration between 2.26 and 9.04 mmol/L). For patients outside this target population, AGE after UC isolation of VLDL is required.
Received July 8, 1999; accepted August 30, 1999.
(1.) Havel RJ. Familial dysbetalipoproteinemia--new aspects of pathogenesis and diagnosis. Med Clin N Am 1982;66:441-54.
(2.) Brewer HB Jr, Zech LA, Gregg RE, Schwartz D, Schaefer EJ. NIH conference. Type III hyperlipoproteinemia. diagnosis, molecular defects, pathology, and treatment. Ann Intern Med 1983;98:623-40.
(3.) Mahley RW, Rail SC Jr. Type III hyperlipoproteinemia (dysbetalipoproteinemia): the role of apolipoprotein E in normal and abnormal lipoprotein metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th ed. New York: McGraw-Hill, 1995:1953-80.
(4.) Durrington PN. Type III hyperlipoproteinemia. In: Hyperlipidaemia-diagnosis and management, 2nd ed. Oxford: Butterworth-Heinemann, 1995:215-24.
(5.) Morganroth J, Levy RI, Fredrickson DS. The biochemical, clinical, and genetic features of type III hyperlipoproteinemia. Ann Intern Med 1975;82:158-74.
(6.) Nakajima K, Saito T, Tamura A, Suzuki M, Nakano T, Adachi M, et al. Cholesterol in remnant-like lipoproteins in human serum using monoclonal anti apoB-100 and anti apoA-I immunoaffinity mixed gel. Clin Chim Acta 1993;53-71.
(7.) Nakajima K, Okazaki M, Tanaka A, Pullinger CR, Wang T, Nakano T, et al. Separation and determination of remnant-like particles in human serum using monoclonal antibodies to apoB-100 and apoA-I. J Clin Ligand Assay 1996;19:177-83.
(8.) Nakajima K, Saito T, Tamura A, Suzuki M, Nakano T, Adachi M, et al. A new approach for the detection of type III hyperlipoproteinemia by RLP-cholesterol assay. J Atheroscler Thromb 1994;1: 30-6.
(9.) Campos E, Nakajima K, Tanaka A, Havel RJ. Properties of an apolipoprotein E-enriched fraction of triglyceride-rich lipoproteins isolated from human blood plasma with a monoclonal antibody to apolipoprotein B-100. J Lipid Res 1992;33:369-80.
(10.) Marcoux C, Tremblay M, Fredenrich A, Jacques H, Krimbou L, Nakajima K, et al. Plasma remnant-like particle lipid and apolipoprotein levels in normolipidemic and hyperlipidemic subjects. Atherosclerosis 1998;139:161-71.
(11.) Marcoux C, Tremblay M, Nakajima K, Davignon J, Cohn JS. Characterization of remnant-like particles isolated by immunoaffinity gel from the plasma of type III and type IV hyperlipoproteinemic patients. J Lipid Res 1999;40:636-47.
(12.) Tomono S, Kawazu S, Kato N, Ono T, Ishii C, Ito Y, et al. Uptake of remnant like particles (RLP) in diabetic subjects from mouse peritoneal macrophages. J Atheroscler Thromb 1994;1:98-102.
(13.) Devaraj S, Vega G, Lange R, Grundy SM, Jialal I. Remnant-like particle cholesterol levels in subjects with dysbetalipoproteinemia or coronary artery disease. Am J Med 1998;104:445-50.
(14.) Leary ET, Wang T, Baker DJ, Cilia DD, Zhong J, Warnick GR, et al. Evaluation of an immunoseparation method for quantitative measurement of remnant-like particle-cholesterol in serum and plasma. Clin Chem 1998;44:2490-8.
(15.) McNamara JR, Shah PK, Nakajima K, Cupples LA, Wilson PWF, Ordovas JM, Schaefer EJ. Remnant lipoprotein cholesterol and triglyceride reference ranges from the Framingham Heart Study. Clin Chem 1998;44:1224-32.
(16.) Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Investig 1955;34:1345-53.
(17.) Hopkins PN, Wu LL, Schumacher MC, Emi M, Hegele RM, Hunt SC, et al. Type III dyslipoproteinemia in patients heterozygous for familial hypercholesterolemia and apolipoprotein E2. Arterioscler Thromb 1991;11:1137-46.
(18.) Havekes LM, de Knijff P, Beisiegel U, Havinga J, Smit M, Klasen E. A rapid micromethod for apolipoprotein E phenotyping directly in serum. J Lipid Res 1987;28:455-63.
(19.) Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with Hhal. J Lipid Res 1990;31:545-8.
(20.) Goldman L. Quantitative aspects of clinical reasoning. In: Wilson JD, Braunwald E, Isselbacher KJ, Petersdort RG, Martin JB, Fauci AS, Root RK, eds. Principles of internal medicine, 12th ed. New York: McGraw-Hill, 1991:5-11.
(21.) Fredrickson DS, Morganroth J, Levy R. Type III hyperlipoproteinemia. an analysis of two contemporary definitions. Ann Intern Med 1975;82:150-7.
TAO WANG,  * KATSUYUKI NAKAJIMA,  ELIZABETH TENG LEARY,  G. RUSS WARNICK,  JEFFREY S. COHN,  PAUL N. HOPKINS,  LILY L. WU,  DONALD D. CILLA,  JIANHUA ZHONG,  and RICHARD J. HAVEL 
 Otsuka America Pharmaceutical, Inc., 2440 Research Blvd., Rockville, MD 20850.
 Pacific Biometrics, Inc. Seattle, WA 98119.
 Clinical Research Institute of Montreal, Montreal, Quebec, Canada H2W 1R7.
 University of Utah, Salt Lake City, UT 84108.
 Cardiovascular Research Institute, University of California, San Francisco, CA 94143.
* Author for correspondence. Fax 301-721-7213; e-mail TAOW@MOCR.OAPI.com.
 Nonstandard abbreviations: HLP, hyperlipoproteinemia; ape, apolipoprotein; TRL, triglyceride-rich lipoprotein; UC, ultracentrifugation; AGE, agarose gel electrophoresis; [beta]-VLDL, [beta]-migrating VLDL; TC, total cholesterol; TG, triglyceride; VLDL-C, VLDL-cholesterol; RLP, remnant-like particle; RLP-C, remnant-like particle-cholesterol; HDL-C, HDL-cholesterol; and LDL-C, LDL-cholesterol.
 All results are presented in SI units. To convert from SI units to "mg/dL", multiply the values by the following conversion factors: 38.61 for cholesterol, 88.50 for triglyceride, and 0.436 for the cholesterol (VLDL-C or RLP-C)-to-total triglyceride ratio. The upper limits of the reference intervals for the ratios of VLDL-C and RLP-C to total triglyceride are 0.69 and 0.23, respectively, for the SI values (or 0.30 and 0.10, respectively, for mg/dL).
Table 1. General characteristics of study subjects. Type III Type IIb (n = 23) (n = 20) Sex, M/F 15/8 12/8 Age, years 52.3 [plusmn] 9.6 50.0 [plusmn] 10.3 Smoker Never 12(52.2%) 10(50.0%) Ex 8(34.8%) 7(35.0%) Current 3(13.0%) 3(15.0%) Alcohol use, Yes/No 12/11 12/8 Hyperthyroidism 1(4.3%) 0(0.0%) Hypothyroidism 3(13.0%) 3(15.0%) Diabetes mellitus 0(0.0%) 0(0.0%) Height, cm 170 [plusmn] 11 166 [plusmn] 11 Weight, kg 91.3 [plusmn] 15.5 86.1 [plusmn] 14.1 BMI,(a) kg/[m.sup.2] 31.3 [plusmn] 6.8 31.4 [plusmn] 6.4 SBP,mmHg 128 [plusmn] 19 130 [plusmn] 16 DBP, mmHg 79 [plusmn] 16 77 [plusmn] 11 Presence of xanthomas 3(13.0%) 3 (15.0%) Type IV (n = 19) Sex, M/F 13/6 Age, years 50.6 [plusmn] 8.3 Smoker Never 13(68.4%) Ex 5(26.3%) Current 1(5.3%) Alcohol use, Yes/No 9/10 Hyperthyroidism 0(0.0%) Hypothyroidism 0(0.0%) Diabetes mellitus 2(10.5%) Height, cm 168 [plusmn] 8 Weight, kg 84.7 [plusmn] 12.6 BMI,(a) kg/[m.sup.2] 29.8 [plusmn] 4.6 SBP,mmHg 130 [plusmn] 17 DBP, mmHg 75 [plusmn] 14 Presence of xanthomas 1 (5.3%) (a) BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure. Table 2. Concentrations (a) of lipids and lipoproteins in the study subjects. Type III Units (n = 23) TC mmol/L 7.12 (6.00, 10.37) mg/dL 275 (232, 401) Total TG mmol/L 4.40 (3.62, 6.86) mg/dL 389 (320, 607) HDL-C mmol/L 0.91 (0.78, 1.04) mg/dL 35 (30, 40) LDL-C mmol/L 2.93 (2.55, 3.48) mg/dL 113 (99, 135) VLDL-C mmol/L 3.26 (2.55, 5.63) mg/dL 126 (99, 218) Estimated [beta]-VLDL-C mmol/L 2.77 (2.09, 5.63) mg/dL 107 (81, 218) RLP-C mmol/L 1.39 (1.04, 2.75) mg/dL 53.5 (40.0, 106.0) Type IIb Type IV (n = 20) (n = 19) TC 7.64 (7.07, 8.65) 5.36 (5.00, 6.24) (b) 295 (273, 334) 207 (193, 241) (b) Total TG 3.40 (2.88, 3.99) (b) 4.01 (3.14, 4.69) (c) 301 (255, 353) (b) 355 (278, 415) (c) HDL-C 0.93 (0.83, 1.06) 0.85 (0.78, 1.10) 36 (32, 41) 36 (32, 41) LDL-C 5.39 (4.77, 6.04) (d) 3.19 (2.85, 3.39) 208 (184, 233) (d) 123 (110, 131) VLDL-C 1.32 (1.04, 1.61) (d) 1.37(l.14, 1.99) (d) 51 (40, 62) (d) 53 (44, 77) (d) Estimated [beta]-VLDL-C 0.04 (0.00, 0.35) (d) 0.00 (0.00, 0.26) (d) 2 (0, 13) (d) 0 (0, 10) (d) RLP-C 0.46 (0.37, 0.63) (d) 0.53 (0.39, 0.76) (d) 18.0 (14.4, 24.5) (d) 20.5 (15.3, 29.5) (d) (a) The results are presented as median (25th, 75th percentile). (b-d) Compared with subjects with type III HLP: (b) P <0.01; (c) P <0.05; (d) P <0.001. Table 3. Diagnostic sensitivity, specificity, and accuracy of various methods compared with the gold standard in diagnosing type III HLP from patients with hypercholesterolemia and with mild to moderate hypertriglyceridemia. Method (cutoff value) Sensitivity (a) Specificity (a) Estimated [beta]-VLDL-C 1.00 0.94 (0.85, 1.00) (>1.04 mmol/L) RLP-C/total TG ([greater than or equal to] 0.23) 0.95 (0.85, 1.00) 0.94 (0.85, 1.00) VLDL-C/total TG ([greater than or equal to] 0.69) 0.80 (0.62, 0.98) 0.97 (0.91, 1.00) Method (cutoff value) Accuracy (a) Estimated [beta]-VLDL-C 0.96 (0.91, 1.00) (>1.04 mmol/L) RLP-C/total TG ([greater than or equal to] 0.23) 0.94 (0.88, 1.00) VLDL-C/total TG ([greater than or equal to] 0.69 0.90 (0.82, 0.98) (a) Values in parentheses are the 95th percentile confidence intervals of the calculated value. If the upper confidence value exceeded 1.00, it was then set at 1.00.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Lipids and Lipoproteins|
|Author:||Wang, Tao; Nakajima, Katsuyuki; Leary, Elizabeth Teng; Warnick, G. Russ; Cohn, Jeffrey S.; Hopkins,|
|Date:||Nov 1, 1999|
|Previous Article:||Dynamic reaction in a homogeneous HDL-cholesterol assay visualized by electron microscopy.|
|Next Article:||Sample preparation and storage can change arsenic speciation in human urine.|