Search

Background: The accuracy of hand-held analytical systems is dependent on the complexity of the instrument and the technical skills of the tester. We evaluated the analytical performance of the LifeScan One Touch II glucose analyzer when used by persons with various levels of technical skill and experience., Methods: A medical technologist conducted tests on three One Touch II systems to ensure that the units were functioning properly. The technologist then trained six analysts (a physician, a registered nurse, a licensed practical nurse, a physician assistant, and two patients) in the proper use of the glucose analyzer., Results: In the hands of a medical technologist, the precision of the three glucose analyzers tested was from 1.0% to 1.9% coefficient of variation (CV) and from 2.3% to 4.2% CV for the low- and high-quality control materials, respectively. The day-to-day precision (5-day period) was from 2.1% to 3.6% CV, and from 3.5% to 4.1% CV for the low- and high-quality control materials, respectively. Comparisons of the glucose values (n = 40 fresh patient serum samples) with a reference method yielded a correlation coefficient of 0.992 to 0.993, and an overall bias of -7%. Although the values obtained by the six operators were statistically different, the differences were not clinically relevant., Conclusions: Our data suggest that the One Touch II glucose analyzer is a reliable system, and that its function is not dependent on the technical skills of the operator.

The objective of this collaborative study with the Department of Veterans Affairs (VA), College of American Pathologists (CAP), and the Centers for Disease Control and Prevention (CDC) was to quantitate the matrix-induced biases of cholesterol measurements on the CAP Comprehensive Chemistry Surveys materials used in proficiency testing (PT). A total of 174 VA Medical Centers outpatient clinics and clinical laboratories participate in the VA-CDC National Cholesterol Standardization and Certification Program. This study was conducted in 112 VA laboratories that have been standardized for measuring cholesterol accurately (within +/- 3.0% of the CDC reference-method values) using fresh, unfrozen, unadulterated human serum samples. Fresh serum samples and 1990 CAP Surveys materials were sent by overnight mail, and the laboratories were asked to analyze them simultaneously in triplicate in a single analytic batch run. The results showed significant matrix-effect biases with the CAP Surveys materials with six of the eight major peer groups, despite the fact that accuracy of cholesterol measurements was maintained with fresh serum samples. The magnitude and direction (positive or negative) of the matrix-effect biases were instrument, reagent, and method specific using the following peer groups: du Pont Dimension (-8.9%); Beckman CX4, CX5, and CX7 (-5.5%); Kodak Ektachem 400, 500, and 700 (+4.4%); Instrumentation Laboratory Monarch (-3.1%); Baxter Paramax (-2.4%); Technicon SMAC and RA (+1.3%); Hitachi/BMD 704 through 747 (+0.4%); and Abbott Spectrum (-0.3%). The CAP PT materials used currently do not behave in a manner identical to fresh human serum when measuring cholesterol on many, but not all, analytic systems. The observed biases due to "matrix effects" with PT materials will cause incorrect conclusions about the accuracy of many laboratory procedures performed on fresh patient specimens. This matrix-effect phenomenon will severely hamper interlaboratory accuracy transfer, standardization efforts, and monitoring performance of a laboratory's testing accuracy with the use of the current survey materials used in PT programs. Collaborative efforts are needed to (1) improve PT fluids to analytically behave more like fresh, human serum; (2) improve instrument design and reagent formulation; and (3) select methods and methodologic parameters that are more "robust" and less sensitive to the exact character of processed calibrators, quality control, and PT materials.

Heightened awareness of the importance of cholesterol and heart disease has increased cholesterol testing in the United States. The demand for reliable cholesterol measurements has become a focal concern of the patient as well as the clinician. This paper covers the major analytical and preanalytical issues and factors that can affect the reliability of cholesterol results. We discuss factors that lead to impression and inaccuracy; solutions for some of the major problems; resources and techniques to help standardize cholesterol measurement; and preanalytical issues that can affect cholesterol results--i.e., patient preparation; collection, processing, storage and proper analysis of the specimen; biological and seasonal variations; age and gender; diet; alcohol consumption; weight changes; exercise; primary diseases; and infections and trauma. Many of these can be controlled by the physician, resulting in more reliable cholesterol readings under stable metabolic conditions. Accurate values will help to classify the patient's coronary heart disease risk, define appropriate treatment strategies, and simplify monitoring of dietary and/or drug intervention.

We describe recent changes in the College of American Pathologists Glycohemoglobin (gHb) Survey, made to improve the assessment of interlaboratory variability and the accuracy of results reported. The questionnaire portion of the survey was revised to include an updated list of current methods, and results for survey specimens were grouped according to the component measured (Hb A1, Hb A1c, or total gHb). The survey specimen material was changed to a material thought to give more reliable results with all available methods. After these changes, instituted in 1989, between-laboratory CVs decreased for some methods. Furthermore, gHb values between method types were more consistent with results obtained from fresh blood samples under very controlled laboratory conditions. However, these recent data also show that the interlaboratory variability is still quite high for some methods and that the variability within and between method types is still very great. We describe a pilot standardization program for gHb measurement.

Many of the difficulties encountered in the original Fiske and SubbaRow method for estimating serum phospholid were corrected through the use of vanadium pentoxide, a catalyst used in sample digestion, and through other minor modifications. Results obtained by the modified method agree well with those obtained by three other established phospholipid procedures. The greatly improved speed, simplicity, accuracy, and precision now makes the modified Fiske and SubbaRos method ideal for clinical or research use.

We describe two cases of atypical dyslipoproteinemia due to drug-induction. This secondary form of lipoprotein abnormality is unique because the newly available drug, miconazole, apparently directly delipidated the alpha-lipoproteins in the bloodstream. On closer study we found that the delipidation was caused by the vehicle rather than the fungicide--more specifically, only by the polyethoxylated castor oil in the vehicle. It affects serum lipoproteins both in vitro and in vivo, and the effect is species-specific. In vitro studies indicate that it preferentially delipidates high-density lipoprotein rather than low-density lipoprotein. Because its effects on the serum lipoproteins of rats resemble those on man, and because aortic lesions were produced in rats injected daily (90 mL/L) with this substance, caution is indicated in long-term use of drugs containing this chemical component in the vehicle.

We describe a method for determining urinary riboflavin by "high-pressure" liquid chromatography, with fluorometric detection. An aliquot of a 24-h urine specimen is injected directly into the chromatograph and the natural fluorescence of riboflavin is measured as the compound is eluted. Interference from other fluorophores is obviated because of differing retention times. Urinary components other than riboflavin and its analogs or degradation products exhibit no significant fluorescence at the wavelengths used (450 nm excitation, 530 nm emission). Analytical recovery of added riboflavin was 96.5 (SD 1.1)%. The CV within-run was 0.7%, between-run it was 4.3%. Concentrations of riboflavin as low as 10 micrograms/L are readily detected, with a linear relation of response to concentration to at least 2000 micrograms/L.

Adult female rats were given daily subcutaneous injections of 17beta-estradiol-3-benzoate, 25 mug/day for twenty days. Rats were then killed under four different experimental conditions: (1) Fasting, killed in the morning (FM), (2) Non-fasting, killed in the morning (NFM), (3) Fasting, killed at night (FN), and (4) Non-fasting, killed at night (NFN). Estradiol induced a three-fold increase in hepatic cholesterol 7alpha-hydroxylase activity in group FM and a significant increase in group NFN. Although increases in 7alhpa-hydroxylase activity might be associated with lower serum cholesterol levels (by increasing the rate of conversion of cholesterol to bile acids), serum cholesterol concentrations were, in fact, elevated in all rats given estradiol.

Recent studies have described an association between high-risk lipoprotein profiles and anabolic steroid abuse by athletes. However, none have included a comprehensive evaluation of diet as a confounding variable. The risk of cardiovascular disease (CVD) and its associations with drug abuse, dietary patterns, and training regimens were evaluated in 18 steroid-using (SU) and 17 non-steroid-using (NSU; no history of drug use or greater than or equal to 1 year drug-free) male bodybuilders. CVD risk was also evaluated in 10 control males. Fasting serum total cholesterol (TC), high-density lipoprotein cholesterol (HDL) and HDL subfractions 2 and 3, low-density (LDL) and very-low-density (VLDL) lipoprotein cholesterol, apoproteins (APO) A-1 and B, and triglycerides (TG) were analyzed at baseline (greater than or equal to 6 months drug-free) and the peak of steroid self-administration in SU. NSU were tested at similar times. Baseline CVD risk factor ratios (TC/HDL) were elevated (greater than 4.97) in 44% of SU and 24% of NSU. When baseline LDL and HDL values were compared to National Cholesterol Education Program CVD risk guidelines, these percentages stayed the same. At the peak of steroid administration significant changes were observed in LDL (22% increase), HDL (63% decrease), HDL-2 (86% decrease), HDL-3 (54% decrease), and TC/HDL (85% increase). No similar measures were observed among NSU or controls. Diets of all bodybuilders were similar, and included a daily intake of 5739 (+/- 2500) kcal, 324 (+/- 163) g protein, 637 (+/- 259) g carbohydrate, 214 (+/- 109) g fat, 5 (+/- 8) g alcohol, 1413 (+/- 1151) mg cholesterol, and a P/S ratio of 0.6 (+/- 0.3). Significant relationships between dietary fats and serum lipids were observed in the NSU. Polyunsaturated fatty acids were correlated with TG and VLDL (r = 0.69; p = 0.01), and TC/HDL (r = 0.06; p = 0.04). Total fats were correlated with TG (r = 0.57; p = 0.05), HDL-3 (r = -0.62; p = 0.04), and VLDL (r = 0.57; p = 0.05), and saturated fats with HDL-3 (r = -0.59; p = 0.055). Diet was moderately associated with lipoproteins in SU, but steroids had a much greater influence on CVD risk. Despite disease promoting diets NSU had relatively average CVD risk that may be attributed to protective effects of rigorous training.

Samples of lyophilized human serum were circulated to more than 7,000 participants in the College of American Pathologists chemistry survey program. The participants measured the concentrations of glucose, cholesterol, uric acid, and urea along with other constituents in the samples. Selected samples also were sent to the National Bureau of Standards (NBS) for analysis of these same analytes by definitive methods. Consensus mean values of participant results are used as target values. The participant values were compared with the NBS values, providing a measure of the accuracy of the designated target values. There was, in general, excellent correlation between the NBS definitive values and the designated target values. These target values can provide reliable reference points by which to evaluate participant performance.

Dietary-induced hypercholesterolemia in swine produces fatty streak lesions in the aortic arch within 15 wks. These lesions, though becoming more extensive in area, do not progress beyond the fatty streak stage up to 30 wks on the diet. The stabilization of these lesions may be attributed, in part, to the extensive migration of foam cells from the lesion into the aortic lumen. Morphological evidence indicates a lumenal movement of these cells. Previous studies (1) indicate that foam cells in these lesions are monocyte-derived. The combined evidence suggests the existence of a monocyte lipid clearance mechanism which may represent an initial response of the blood-vascular system to atherogenesis.

The National Cholesterol Education Program has recommended that all laboratories be consistent, precise, and accurate in the reporting and measurement of blood cholesterol levels. In a follow-up to a 1984 survey study, we assessed the changes in reporting procedures for measurements of blood lipid levels in 16 clinical laboratories in Nebraska. Using human serum reference materials of known cholesterol concentrations provided by the Centers for Disease Control, we also assessed the precision and accuracy of measurement of blood cholesterol levels in clinical laboratories in Nebraska. Fourteen of the 16 laboratories restudied in 1987 had altered the reference range for total serum cholesterol since 1984, 86% of whom lowered the upper limit of the reference range. Eleven of 16 laboratories expressed reference ranges for total serum cholesterol in terms of patient age in 1987, while only 7 of 20 did in 1984. Gender-based reference ranges increased from 0 to 5 from 1984 to 1987. Similar trends were seen in the reporting of high-density lipoprotein cholesterol and triglyceride concentrations. Reporting procedures varied greatly; only 1 laboratory used National Cholesterol Education Program risk levels for measuring total serum cholesterol levels. Fifteen laboratories met the National Cholesterol Education Program recommendation for precision (coefficient of variation, less than or equal to 5%) and 78% of laboratories obtained results that satisfied the current recommendation for accuracy (within 5% of "true value," as determined by the Centers for Disease Control).

We have approached a dawn of a new era in detection, evaluation, treatment, and monitoring of individuals with elevated blood cholesterol levels who are at increased risk for CHD. The NHLBI's National Cholesterol Education Program will be the major force underlying this national awareness program, which is dependent on the clinical laboratories providing reliable data. Precision or reproducibility of results is not a problem for most of the laboratories, but accuracy is a major concern. Both the manufacturers and laboratorians need to standardize the measurement for cholesterol so that the accuracy base is traceable to the NCCLS NRS/CHOL. The manufacturers need to adopt a uniform policy that will ensure that the values assigned to calibration, quality control, and quality assurance or survey materials are accurate and traceable to the NCCLS/CHOL. Since, at present, there are some limitations of these materials caused by matrix effects, laboratories are encouraged to use the CDC-NHLBI National Reference Laboratory Network to evaluate and monitor their ability to measure patient blood cholesterol levels accurately. Major areas of analytical problems are identified and general, as well as specific, recommendations are provided to help ensure reliable measurement of cholesterol in patient specimens.

A semipurified diet containing 43% of the calories as fat and 1.2 mg of cholesterol/cal was fed to cynomolgus monkeys (Macaca fascicularis) for 6 months; the cholesterol content was reduced to 0.34 mg/cal for the next 18 months. During the latter period, the monkeys were assigned to 4 groups of 18 animals each and received the following dietary additions: A, none (controls); B, cholestyramine (5%, w/w); C, dextrothyroxine (0.003%); and D, Wy-14,643 (0.45%). Cholestyramine normalized plasma lipid levels and reduced the size of aortic and coronary atherosclerotic lesions in spite of the high-fat, high-cholesterol intake. Dextrothyroxine reduced cholesterolemia but did not modify the extent of arterial lesions. Wy-14,643 changed neither plasma cholesterol levels nor the extent of atherosclerosis.

An unusual serum lipoprotein (Lp) profile was detected in a Japanese family. A double beta-Lp was observed when serum was subjected to polyacrylamide gel electrophoresis. The slower migrating beta-Lp was identified as a subfraction of high density lipoprotein (HDL). It was present in the d 1.063--1.21 fraction, migrated to the position designated as the midband L.1 (sinking pre-beta-lipoprotein) by Mead, M.G. and Dangerfield, W.G. (1974) (Clin. Chim. Acta 51, 173--182) [1], and reacted against human anti-beta-Lp antiserum. This lipoprotein contained greater amounts of triglyceride than the usual beta-lipoprotein and could not be clearly detected by paper electrophoresis. Individuals exhibiting this high density midband lipoprotein appeared to be heterozygous for an autosomal dominant gene. Although other reports have indicated the possibility of a positive association between the occurrence of serum lipoproteins with unusual eletrophoretic mobility and premature ischemic heart disease, no such correlation was demonstrable in these subjects.

In a previous publication (1) we demonstrated that atherosclerotic lesion development in the aorta of hypercholesterolemic pigs was preceded by intimal penetration by blood-borne mononuclear cells, and that medial smooth muscle cells were not involved in the development of fatty streak lesions in this model. The current study shows that lesions in the aortic arch do not progress beyond the fatty streak stage up to 30 wk on a hypercholesterolemic diet, and that mononuclear cell involvement continues during this period. Mononuclear cells adherent to the endothelium, or in the intima of lesion areas were identified as monocytes by the presence of peroxidase-positive granules (peroxisomes) in their cytoplasm. As well, "modified" intimal monocytes which contained a few lipid droplets and which were phagocytic, also sometimes contained peroxisomes. The results suggest that monocytes which invade the intima before or during lesion development undergo transformation into macrophage foam cells in fatty streak lesions. The absence of medial cell involvement indicates that monocytes are the major foam cell precursor in these lesions.

An electroimmunoassay has been developed to quantify apolipoprotein B (apoB) in samples of aorta from normolipemic and hypercholesterolemic (diet-induced) swine. Acceptability of the assay was demonstrated using tests for specificity, sensitivity, accuracy, and validity. Specific sites in the grossly normal aortic arch of hypercholesterolemic animals were macroscopically demarcated after the injection of the dye, Evans blue. These blue sites, previously shown to be regions of early lesion formation, contained significantly greater amounts of apoB than adjacent nonblue (white) zones. By contrast, in normolipemic swine no difference in apoB content was found between blue and white regions of the arch. Likewise, the apoB content did not differ among various white regions of grossly normal aortas of hypercholesterolemic swine when no lesions were present. Also, once gross lesions were found in white regions, the apoB content was at least as great as in blue regions. The differences in apoB content between lesioned and nonlesioned areas were greatest in the abdominal aorta. The results of this study are consistent with an enhanced interstitial accumulation of lipoproteins containing apoB, especially low-density lipoprotein, occurring prior to lesion formation in this experimental model.

A sensitive, reproducible, and specific method for the determination of urinary thiamine has been established. Unique to the use of high-pressure liquid chromatography (HPLC) to separate the fluorescent thiamine derivative from interfering fluorescent compounds. Urine samples were passed through a Decalso catoin-exchange column, washed with 0.5 M KCl to remove some interfering compounds, and eluted with 3.4 M KCl. The eluted thiamine was converted to the fluorescent derivative, thiochrome, by reaction with alkaline potassium ferricyanide. The reaction mixture was extracted with isobutanol and subjected to HPLC monitored by a fluorescent detector. Within-day and day-to-day coefficients of variation proved to be 2.5% and 1.2% respectively. Recovery of added thiamine (range 0.04 to 2.0 microgram/ml) averaged 99.9 +/- 5.3%. The sensitivity of this method was 0.03 microgram/ml.

The National Heart, Lung, and Blood Institute national awareness program on cholesterol and heart disease has placed new demands on laboratorians to utilize and perform more reliable measurements of lipids, lipoproteins, and apolipoproteins. The general public's awareness and the clinicians' concerns about the reliability of laboratory testing make it paramount that the analytical problems and issues are identified and solutions are provided to increase the current state of reliability of the measurement of these blood constituents. To accomplish this, the initial step is to assess the current state of reliability of lipid, lipoprotein, and apolipoprotein measurements in the clinical laboratories. Accuracy and precision of measurements of total cholesterol, triglycerides, high-density lipoprotein cholesterol, and apolipoproteins A-I and B are extensively discussed, and general as well as some specific recommendations are provided for some of the apparent problems.

Hypercholesterolemia was induced in pigs by feeding a chow diet supplemented with 1.5% cholesterol and 19.5% lard for periods up to 12 weeks. The aortic intima from areas of spontaneously differing permeability to proteins, as demarcated by their uptake of Evans blue dye, was examined using light microscopy and both scanning and transmission electron microscopy to describe the earliest detectable changes in intimal morphology induced by the diet. After 2, 4, and 6 weeks of feeding, cholesterol/lardfed pigs demonstrated monocyte adherence to the endothelium in areas of enhanced permeability (blue areas) in 86% of samples examined, as compared to 52% in areas of lesser permeability (white areas) and 17% in control animals. Similarly, the number of monocytes in the intima was higher in blue areas than in adjacent white areas or blue areas from control animals. After 12 weeks of feeding, all blue areas showed intimal monocytes, with fewer seen in white areas. Aortic endothelial cells in hypercholesterolemic pigs were normal in ultrastructural appearance, except they contained more lysosomes and cytoplasmic filaments than those from control animals. No lesions were observed at 2, 4, and 6 weeks, although plasma cholesterol levels were substantially elevated (200-400 mg/dl) at these times. A marked hyper-beta-lipoproteinemia was evident from 4 weeks onward, but no elevation of serum triglycerides was evident at any stage. Plasma phospholipid concentrations increased but not in direct proportion to cholesterol levels. At 12 weeks, foam cell lesions were observed in areas of enhanced permeability but not in adjacent areas of normal permeability. Lesion foam cells appeared to be derived from the monocytes which adhered to and penetrated the endothelium at earlier stages, since no intimal involvement, or lipid engorgement, by medial smooth muscle cells was observed.

A semipurified diet containing 1.2 mg of cholesterol/Cal was fed to cynomolgus monkeys (Macaca fascicularis). At the end of 6 months, a group of 18 animals was killed for evaluation of atherosclerosis in the aorta and the coronary arteries. The remaining monkeys were assigned to three groups of 18 animals each and fed, during the following 18 months, semipurified diets containing 0.34 mg of cholesterol/Cal with or without alfalfa meal, or a diet consisting entirely of Monkey Chow. a decrease in cholesterolemia and plasma phospholipid levels, normalization in the distribution of plasma lipoproteins, and reduction in the extent of aortic and coronary atherosclerosis were observed in monkeys fed the semipurified diet containing alfalfa, although the intake of cholesterol remained as high as in the usual American diet. These changes, also observed in monkeys fed a chow diet almost devoid of cholesterol, suggest that alfalfa counteracts the atherogenic effect of dietary cholesterol.

Past studies on urinary loss of magnesium (Mg) have focused on young diabetic rats. The aims of the present study were to determine the rapidity at which glycosuria and magnesuria occur after the induction of diabetes mellitus (DM) in old male rats, and the maximal amount of Mg and glucose (Glu) loss in the urine and whether or not the loss is persistent and (3) the most sensitive means of correlating the Mg and Glu loss in the urine. Three methods of expressing urinary Mg and Glu concentrations were selected: Method A: mg/24 h; Method B: mg/24 h/kg body weight (BW), and Method C: ratio of Mg to creatinine (Mg/Crea) or ratio of Glu to Crea (Glu/Crea). Our study indicated a maximal and rapid loss of Mg and Glu occurring within 1 week after induction of DM (by streptozotocin injection) and remained in effect for 6 weeks, the end of the study. The urinary Mg loss due to DM correlated best with urinary Glu when expressed as mg/24 h/kg BW. In addition, the increase in urinary Mg concentration paralleled the degree of glycosuria and reached a maximum of 5-12 times baseline values when expressed by Method B. This was 4-10 times when expressed by Method A, and 2-6 times when expressed by Method C. Since polyuria is a feature of DM, we also correlated the relationship between these two factors, and found a significantly positive correlation (r = 0.735) particularly when Mg was expressed as mg/24 h. In summary, rapid and significant urinary Mg loss is observed in old rats made diabetic with streptozotocin.(ABSTRACT TRUNCATED AT 250 WORDS)

We studied 226 adult male subjects (mean=52yr) who underwent coronary arteriography. Their serum lipid and lipoprotein levels [total cholesterol (TC), triglycerides (TG), HDL-cholesterol (HDL-C), LDL-C, HDL-phospholipid (HDL-P), HDL-TG, HDL-C/TC, HDL-P/TC, LDL-C/TC, LDL-C/HDL-C and LDL-C/HDL-P] were correlated with the severity of coronary artery disease (CAD). Studies showed a slight but statistically significant correlation (linear regression) between CAD and HDL-P/TC (r=-0.24, P < 0.001), HDL-C/TC (r=-0.20, P < 0.002), followed by LDL-C/HDL-P (r=0.19, P < 0.004) and HDL-P (r=-0.18, P < 0.008). HDL-C, LDL-C and TC were significant at the P < 0.05 level; P value of TG was non-significant. Subjects were conveniently grouped based on degree of coronary artery narrowing: normal, mild (1-50%); moderate-severe (51-99%) and very severe (100% occlusion). Of the 11 lipid variables, the best predictor of the stage of the CAD was HDL-P/TC as measured by one-way analysis of variance. This trend was unaltered even after adjustment for covariates. HDL-P, LDL-C/HDL-P and HDL-C/TC were also significant, but the other lipid parameters were not. The study indicates that HDL-C, by itself, is not as effective a predictor of CAD as HDL-P/TC. Also, the small but statistically significant inverse relationship between HDL-P/TC and CAD suggests that a low HDL-P/TC ratio can be considered a risk factor for CAD but not as a dependable clinical diagnostic aid for predicting the severity of CAD on an individual basis.

We carefully selected 30 men with primary gout, rendered asymptomatic by therapy, to examine the frequency and type of hyperlipidemia and hyperlipoproteinemia, with the objective of determining whether serum uric acid, alcohol intake, liver function, kidney function, and (or) drugs were participating in the secondary lipid disorder. Sixty-one age- and sex-matched men were used as controls. About 73% of the gout patients had hypertriglyceridemia, 1.6-fold the frequency found in the control group. Types IV and IIb lipoprotein electrophoretic patterns were most prevalent in the gout group. Neither alcohol intake nor hyperuricemia, per se, seems to be the cause of the lipid and lipoprotein disorder and cannot be related to liver or kidney dysfunctions. Obesity was the major underlying factor associated with the lipidemia. The study suggests that diet and, possibly, defective clearance of triglycerides may be etiologic factors associated with the abnormal serum triacylglycerol (triglyceride) and lipoprotein concentrations in these individuals.

The observations so far conducted in cynomologus monkeys on semipurified diets containing butter and cholesterol suggest that this nonhuman primate is an excellent model for studying the therapy of established coronary atherosclerosis. (1) This species is available at a reasonable cost and can be kept in captivity in good health for prolonged periods of time. (2) It readily accepts semipurified diets with a percentage composition similar to that of human diets in the U.S. (3) Ingestion of these diets leads quite rapidly (around 6 months) to moderate coronary atherosclerosis. More prolonged feeding leads to lesions which are histologically very similar to those in man. (4) The distribution of lesions in the main coronary arteries is similar to that in man. (5) Methods to quantify the coronary lesions are available. (6) The diets can be so modified that cholesterol levels closely resemble those in hypercholesterolemic man. (7) The monkeys are amenable to several therapeutic regimens which show promise of arresting the progress or inducing the regression of the coronary lesions.