(Circulation. 1996;94:2351-2354.)
© 1996 American Heart Association, Inc.
Articles |
Beyond LDL Cholesterol Reduction
Cholesterol, Genetics, and Heart Disease Institute and Berkeley HeartLab, San Mateo, and Lawrence Berkeley National Laboratory, Berkeley, Calif.
Correspondence to H. Robert Superko, MD, Cholesterol, Genetics, and Heart Disease Institute, 1875 S Grant St, Suite 700, San Mateo, CA 94402. E-mail superko@best.com. http://www.heartdisease.org.
Key Words: Editorials • cholesterol • lipoproteins
 |
Success of LDL-C Reduction |
|---|
 Top Success of LDL-C Reduction Failure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
Within the past decade, clinical trials of LDL-C reduction have convincingly demonstrated that LDL-C reduction in primary and secondary prevention trials can significantly reduce clinical cardiac events.1 Arteriographic investigations have demonstrated that LDL-C reduction can significantly reduce the rate of arteriographically defined disease progression.1
|
Failure of LDL-C Reduction |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
Despite the success of LDL-C reduction, close examination of the trial results reveals that a substantial number of subjects who received treatment and achieved significant LDL-C reduction still had a clinical event or evidence of arteriographic progression (Table 1
). In the LRC-CPPT, for example, there was a 17% reduction in clinical events, which was made up of 187 events in the control group and 155 in the treatment group (32 fewer events). In SSSS, there was a 30% reduction in clinical events, which was composed of 622 events in the placebo group and 431 in the treatment group (191 fewer events). Although the reduction in clinical events is gratifying and laudable, it was not enough for the 155 subjects in the LRC-CPPT or the 431 subjects in SSSS who received treatment yet still had an event. The reason for such a large number of poor responders may lie in the prevalence of metabolic abnormalities linked to atherosclerosis that are not detected on routine laboratory tests and hence are not adequately treated with just LDL-C reduction.2 3 This issue also involves the concept of monotherapy, which takes on added importance with the report that multifactorial risk reduction reduces clinical events significantly more than single therapy.4
|
These disorders have a strong familial inheritance pattern, and CAD can be considered a genetic disease attributable to multiple gene-environment interactions.5 The Figure illustrates the approximate prevalence of well-recognized lipoprotein disorders along with disorders not routinely screened for. The well-established disorders familial heterozygous hypercholesterolemia, familial combined hyperlipidemia, and hypoalphalipoproteinemia can be detected in 
3% to 15% of CAD patients. Other disorders, such as apoprotein E isoform differences, hyperapobetalipoproteinemia, homocysteinemia, ALP disorder, and Lp(a), can be detected in 30% to 50% of male CAD patients.3
|
|
Lp(a) and the Laboratory Problem |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
The evidence that elevated Lp(a), particularly in the presence of other risk factors, is useful in predicting CAD risk is substantial.6 Knowledge of a patient's Lp(a) value is of particular use in predicting atherosclerosis risk when other risk factors, such as high LDL-C, are present.7 Although knowledge about the effect of Lp(a) reduction is sparse, it is fast becoming a test requested by physicians dealing with atherosclerosis. One of the problems that argues against the routine use of a test for Lp(a) concentration involves laboratory issues of precision, accuracy, and quality control. In response to a local physician's inquiry regarding a large discrepancy between Lp(a) measurements for one patient determined in a commercial laboratory, we conducted a small comparison study of laboratories offering this service (Table 2
). Samples were drawn from nine stable subjects for Lp(a) analysis on two occasions 1 week apart and were sent to the LBNL and three different commercial laboratories offering the test. At the LBNL, the mean difference between tests was 2.7±2.7 mg/dL, with a maximum error of +5 and a minimum error of -1. The commercial laboratories had either a larger mean difference or a wide maximum-minimum range. This survey suggests a wide variation in results for commercially available measurement of Lp(a) and brings into question the clinical use of a test with such wide measurement variation. One difficulty in Lp(a) interpretation involves laboratory issues of quality control. There are several methods of detecting Lp(a), including radioimmunoassay, radial immunodiffusion, rocket immunoelectrophoresis, and ELISA.8 The simple act of using different reagents or kits can result in significant measurement variability, which has previously been shown to be a problem with HDL-C measurement as well.9 Furthermore, there is no universally accepted standard, and even if all laboratories used the same reagents and protocol, wide variation would be seen because of the lack of a universal standard. Finally, Lp(a) is an acute-phase reactant, and some variability may be physiological. Knowledge of a patient's Lp(a) value can be clinically useful; however, performing this analysis with accuracy, precision, and reproducibility is difficult, and inaccurate results may result in inappropriate clinical decisions. If clinicians wish to use this test, it would be wise to obtain it from a laboratory linked to published research studies that indicate population values and research quality test results.
|
|
Atherogenic Lipoprotein Profile |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
Although all the disorders in the figure have been linked to CAD risk, one in particular, ALP, has significant basic science and clinical trial support indicating that it is a useful clinical tool. The importance of lipoprotein subclasses and triglyceride-rich lipoprotein particles is not new; it dates back 30 years to the groundbreaking work of Gofman et al presented at the 1965 Lyman Duff Memorial Lecture.10 Serum samples from men in the Framingham Study analyzed at LBNL revealed significantly higher (P<.001) LDL mass (Sf 0 to 12), IDL (Sf 12 to 20), and triglyceride-rich particles in the Sf 20 to 100 and Sf 100 to 400 ranges in subjects who developed CAD compared with those who did not. Because of correlation between several of the variables, it was not possible to translate the positive findings into any statement of independent contribution. The results of the Livermore Study were consistent with these results and indicated that lipoprotein classes other than LDL, particularly triglyceride-rich, more buoyant particles, contribute significantly to CAD risk.
Since 1965, a substantial body of knowledge has become available regarding specific biochemical and metabolic features of the IDL, LDL, and HDL subpopulations that relate to atherogenesis.11 12 The ALP, which is characterized by a predominance of small, dense LDL particles, is also associated with reduced levels of HDL2, increased postprandial lipemia, LDL particles susceptible to oxidation due to reduced vitamin E content, enhanced arterial wall uptake, and insulin resistance. Thus, the ALP trait, also called LDL subclass pattern B, is a heritable trait composed of several metabolic disorders that result in an atherogenic metabolic stew. This body of knowledge was recently acknowledged by the 27th Bethesda Conference of the American College of Cardiology, Task Force 4.13
|
ALP: The Clinical Evidence |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
The CAD risk associated with ALP has been demonstrated in the Boston Area Heart Health Project and most recently in the prospective Physicians' Health Survey and the Stanford Five City Project, in which the LDL pattern B was associated with a threefold increased CAD risk independent of many of the classic CAD risk factors, including total cholesterol, HDL-C, body mass index, and apolipoprotein B.14 15 In the Stanford Five City Project, LDL size was reported to be the strongest physiological risk factor in conditional logistic regression. ALP has many of the metabolic characteristics of non–insulin-dependent diabetes mellitus, and indeed, the small LDL pattern B trait has also been shown to be a risk factor for the future development of non–insulin-dependent diabetes mellitus and implies that the small LDL trait contributes to the risk of CAD in prediabetic subjects.16
Over the past several years the importance of lipoprotein subclasses for arteriographic determinants of atherosclerosis has been clarified. In a nonhypercholesterolemic CAD population, the natural history of arteriographic progression and clinical events was significantly correlated with IDL and HDL but not LDL.17 The characteristics of elevated IDL and reduced HDL are found in LDL subclass pattern B subjects. The NHLBI-II trial reported in 1987 that subjects classified as showing arteriographic "stability" had significantly greater reductions in total LDL mass (Sf 0 to 12), small LDL (Sf 0 to 7), and IDL (Sf 12 to 20).18 The MARS trial reported that in patients successfully treated with a statin and achieving an LDL-C <85 mg/dL, triglyceride-rich lipoprotein levels were the predominant predictor of progression.19 The STARS investigation reported that "dense" LDL was reduced significantly more in the group classified as showing arteriographic regression than in the progression group and that reduction in dense LDL was the best predictor of arteriographic outcome.20 In a recent Circulation article, the Stanford Coronary Risk Intervention Project reported that despite almost identical LDL-C reduction in patients with predominantly dense (probable pattern B) or buoyant (probable pattern A) LDL particles, there was no significant arteriographic difference between treatment and control pattern A subjects, whereas a significant reduction in the rate of arteriographic progression was seen in the treatment versus control dense LDL (probable pattern B) subjects.21 This body of knowledge reflects the importance of lipoprotein subclasses in the atherogenic process that was first raised by Gofman and colleagues.10
|
ALP: Differential Response to Treatment |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
Individual variability in response to lipid treatment has been clinically observed for many years. The presence of LDL subclass pattern A or B explains a portion of this variability. The importance of this issue is exemplified by the Helsinki Trial, which reported that a subset of subjects characterized by moderate elevations in triglycerides and characteristics of the insulin-resistance syndrome, which is similar to the small-LDL trait, received the majority of the clinical event reduction benefit from gemfibrozil treatment.22 23 Likewise, this indicates that subjects without these traits received little clinical event benefit from gemfibrozil treatment. Recently, in a double-blind, randomized trial, gemfibrozil has been reported to have little to no effect on total LDL mass in LDL pattern A and B individuals.24 In LDL pattern B subjects, however, there was a significant reduction in small LDL counterbalanced by a significant increase in large LDL, whereas LDL pattern A subjects revealed no gemfibrozil effect in either large or small LDLs. A similar differential response to therapy between LDL pattern A and B subjects has been reported with nicotinic acid, bile acid–binding resin, and reduced-fat diet therapy.25 26 27 Thus, common lipid-altering therapies can have significant effects on lipoprotein subclass distributions that are not revealed by routine tests of lipoprotein cholesterol. These differences in response can guide the clinician toward choosing the most efficacious treatment for individual patients.
The importance of triglyceride-rich lipoprotein particles has now been established at the basic science level, in animal studies, in cross-sectional and prospective human trials, and in arteriographic "regression" trials (Table 3). Measures of these particles provide diagnostic and response-to-treatment information that is not available through routine lipid panels. This abundance of evidence now makes the application of this knowledge on a clinical level attractive, because it refines diagnostic and therapeutic abilities and provides improved health care.
|
|
One of the Remaining Problems |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
A remaining problem involves the laboratory aspects of measuring triglyceride-rich lipoproteins or lipoprotein subclass distribution. One method for determining the presence of triglyceride-rich particles is by determination of apoproteins associated with these particles. However, only a few large clinical trials have used this approach.19 28 The presence of large or small LDL and IDL and VLDL mass in many of the investigations mentioned above has been determined by ANUC at the LBNL. This is the same method as that pioneered by Drs John Gofman and Frank Lindgren. Although it provides tremendous detail, its clinical use is hindered by the expense and the expertise required to maintain and operate the system. Other techniques are available, including density gradient ultracentrifugation and GGE. Of these, GGE has been used in the largest number of clinical trials and can provide detailed information on LDL particle size and percent distribution in seven LDL subclasses.29 The GGE method at the LBNL also benefits from the ability to compare both ANUC and GGE results obtained simultaneously in numerous clinical trials. Thus, a cost-effective correlation between ANUC results and GGE results can be obtained, and lessons learned from these clinical trials can be applied to patient management. A problem with the GGE methodology was created several years ago, when the gels used most often by researchers in the field were no longer commercially available. This required that research institutions develop their own gel production methods, which provide adequate supplies for research purposes, but a large commercial supply of gels proven to provide adequate information appears to be lacking. In addition, although the percent distribution within LDL subclasses can be reproducibly determined, quantification of cholesterol content in subclasses is not currently available. Future tests should include quantitative subclass distribution as well as quantification of IDL particles.
Although ongoing and future investigations will add knowledge to our understanding of lipoprotein subclasses and CAD, enough evidence exists to use our present knowledge to provide a more refined approach to atherosclerosis risk determination and treatment with our current tools. Elevated LDL-C is only one of several metabolic disorders contributing to CAD risk. The importance of the small-LDL pattern B trait and triglyceride-rich lipoproteins probably exceeds that of LDL-C, because more CAD patients are found to have the LDL pattern B trait than hypercholesterolemia. Application of this knowledge assists in creating the optimal metabolic milieu to encourage atherosclerosis stability, provides clinicians with additional tools with which to make therapeutic decisions, and affects cost-effectiveness by matching the disorder to the most appropriate therapy. Each day we wait, more lives are damaged. Now is the time to implement these changes. As Mark Twain was reputed to have said, "You may be on the right track, but, if you just sit there, you'll get run over."
|
Selected Abbreviations and Acronyms |
|---|
|
|
Footnotes |
|---|
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
|
TopSuccess of LDL-C ReductionFailure of LDL-C ReductionLp(a) and the Laboratory...Atherogenic Lipoprotein ProfileALP: The Clinical EvidenceALP: Differential Response to...One of the Remaining...References |
|---|
1. Superko HR, Krauss RM. Coronary artery disease regression: convincing evidence for the benefit of aggressive lipoprotein management. Circulation.. 1994;90:1056-1069.
2.
Genest JJ, Martin-Munley SS, McNamara JR, Ordovas JM, Jenner J, Meyers RH, Silberman SR, Wilson PWF, Salem DN, Schaefer EJ. Familial lipoprotein disorders in patients with premature coronary artery disease. Circulation.. 1992;85:2025-2033.
3. Superko HR. New aspects of cardiovascular risk factors including small, dense LDL, homocysteinemia, and Lp(a). Curr Opin Cardiol.. 1995;10:347-354.[Medline] [Order article via Infotrieve]
4. Thompson GR, Hollyer J, Waters DD. Percentage change rather than plasma level of LDL-cholesterol determines therapeutic response in coronary heart disease. Curr Opin Lipidol.. 1995;6:386-388.[Medline] [Order article via Infotrieve]
5. Superko HR. Inherited disorders in the lipoprotein system: a common cause of premature heart disease. In: Julian DG, Wenger NK, eds. Women and Heart Disease. London, UK: Martin Dunitz Ltd; 1996.
6. Scanu AM, Gless GM. Lipoprotein (a): heterogeneity and biological relevance. J Clin Invest.. 1990;85:1709-1715.
7.
Schaefer EJ, Lamon-Fava S, Jenner JL, McNamara JR, Ordovas JM, Davis E, Abolafia JM, Lippel K, Levy RI. Lipoprotein(a) levels and risk of coronary heart disease in men. JAMA.. 1994;271:999-1003.
8. Fless GM, Snyder ML, Scanu AM. Enzyme-linked immunoassay for Lp(a). J Lipid Res.. 1989;30:651-662.[Abstract]
9.
Superko HR, Bachorik PS, Wood PD. High-density lipoprotein cholesterol measurements. JAMA.. 1986;256:2714-2717.
10.
Gofman JW, Young W, Tandy R. Ischemic heart disease, atherosclerosis, and longevity. Circulation.. 1966;34:679-697.
11. Krauss RM. Heterogeneity of plasma low-density lipoproteins and atherosclerosis risk. Curr Opin Lipidol.. 1994;5:339-349.[Medline] [Order article via Infotrieve]
12.
Slyper AH. Low-density lipoprotein density and atherosclerosis: unraveling the connection. JAMA.. 1994;272:305-308.
13. Forrester JS, Merz NB, Bush TL, Cohn JN, Hunninghake DB, Parthasarathy S, Superko HR. Task Force 4: efficacy of risk factor management. J Am Coll Cardiol.. 1996;27:991-1006.[Medline] [Order article via Infotrieve]
14. Stampfer MJ, Krauss RM, Blanche PJ, Holl LG, Sacks FM, Hennekens CH. A prospective study of triglyceride level, low density lipoprotein particle diameter, and risk of myocardial infarction. JAMA. 1996;882-888.
15.
Gardner CD, Fortmann SP, Krauss RM. Small low density lipoprotein particles are associated with the incidence of coronary artery disease in men and women. JAMA.. 1996;276:875-881.
16.
Austin MA, Mykkanen L, Kuusisto J, Edwards KL, Nelson C, Haffner SM, Pyorala K, Laakso M. Prospective study of small LDLs as a risk factor for non–insulin dependent diabetes mellitus in elderly men and women. Circulation.. 1995;92:1770-1778.
17.
Phillips NR, Waters D, Havel RJ. Plasma lipoproteins and progression of coronary artery disease evaluated by angiography and clinical events. Circulation.. 1993;88:2762-2770.
18. Krauss RM, Lindgren FT, Williams PT, Kelsey SF, Brensike J, Vranizan K, Detre KM, Levy RI. Intermediate-density lipoproteins and progression of coronary artery disease in hypercholesterolaemic men. Lancet.. 1987;2:62-65.[Medline] [Order article via Infotrieve]
19.
Hodis HN, Mack WJ, Azen SP, Alaupovic P, Pogoda JM, LaBree L, Hemphill LC, Kramsch DM, Blankenhorn DH. Triglyceride- and cholesterol-rich lipoproteins have a differential effect on mild/moderate and severe lesion progression as assessed by quantitative coronary angiography in a controlled trial of lovastatin. Circulation.. 1994;90:42-49.
20. Watts GF, Mandalia S, Brunt JN, Slavin BM, Coltart DJ, Lewis B. Independent associations between plasma lipoprotein subfraction levels and the course of coronary artery disease in the St Thomas' Atherosclerosis Regression Study (STARS). Metabolism.. 1993;42:1461-1467.[Medline] [Order article via Infotrieve]
21.
Miller BD, Alderman EL, Haskell WL, Fair JM, Krauss RM. Predominance of dense low-density lipoprotein particles predicts angiographic benefit of therapy in Stanford Coronary Risk Intervention Project. Circulation.. 1996;94:2146-2153.
22.
Manninen V, Tenkanen L, Koskinen P, Huttunen JK, Manttari M, Heinonen OP, Frick MH. Joint effects of serum triglyceride and LDL cholesterol and HDL cholesterol concentrations on coronary heart disease risk in the Helsinki Heart Study. Circulation.. 1992;85:37-45.
23.
Tenkanen L, Manttari M, Manninen V. Some coronary risk factors related to the insulin resistance syndrome and treatment with gemfibrozil. Circulation.. 1995;92:1779-1785.
24. Superko HR, Krauss RM. Reduction of small, dense LDL by gemfibrozil in LDL subclass pattern B. Circulation. 1995;92(suppl I):I-250. Abstract.
25. Superko HR, and the KOS Investigators. Effect of nicotinic acid on LDL subclass patterns. Circulation. 1994;90(suppl I):I-504. Abstract.
26. Superko HR, Williams PT, Alderman EL, and the Stanford Coronary Risk Intervention Project Investigators. Differential lipoprotein effects of bile acid binding resin in LDL subclass pattern A versus B. Circulation. 1992;86(suppl I):I-144. Abstract.
27. Dreon DM, Fernstrom H, Miller B, Krauss RM. Low density lipoprotein subclass patterns and lipoprotein response to a reduced-fat diet in men. FASEB J.. 1994;8:121-126.[Abstract]
28.
Blankenhorn DH, Alaupovic P, Wickham E, Chin HP, Azen SP. Prediction of angiographic change in native human coronary arteries and aortocoronary bypass grafts. Circulation.. 1990;81:470-476.
29. Krauss RM. Heterogeneity of plasma low-density lipoproteins and atherosclerosis risk. Curr Opin Lipidol.. 1994;5:339-349.
This article has been cited by other articles:
 |
|
 |
|
  S. Caron and B. Staels Apolipoprotein CIII: A Link Between Hypertriglyceridemia and Vascular Dysfunction? Circ. Res., December 5, 2008; 103(12): 1348 - 1350. [Full Text] [PDF]
|
|
|
|
 |
|
 H. R. Superko and S. King III Lipid Management to Reduce Cardiovascular Risk: A New Strategy Is Required Circulation, January 29, 2008; 117(4): 560 - 568. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zambon, P. Gervois, P. Pauletto, J.-C. Fruchart, and B. Staels Modulation of Hepatic Inflammatory Risk Markers of Cardiovascular Diseases by PPAR-{alpha} Activators: Clinical and Experimental Evidence Arterioscler Thromb Vasc Biol, May 1, 2006; 26(5): 977 - 986. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. R. van Lennep, H. T. Westerveld, H. W. O. R. van Lennep, A. H. Zwinderman, D. W. Erkelens, and E. E. van der Wall Apolipoprotein Concentrations During Treatment and Recurrent Coronary Artery Disease Events Arterioscler Thromb Vasc Biol, November 1, 2000; 20(11): 2408 - 2413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Avins and J. M. Neuhaus Do Triglycerides Provide Meaningful Information About Heart Disease Risk? Arch Intern Med, July 10, 2000; 160(13): 1937 - 1944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rubenfire Update in Cardiology Ann Intern Med, September 7, 1999; 131(5): 356 - 362. [Full Text] [PDF]
|
|
|
|
|
|
A. Zambon, J. E. Hokanson, B. G. Brown, and J. D. Brunzell Evidence for a New Pathophysiological Mechanism for Coronary Artery Disease Regression : Hepatic Lipase–Mediated Changes in LDL Density Circulation, April 20, 1999; 99(15): 1959 - 1964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Ballantyne, J. A. Herd, L. L. Ferlic, J. K. Dunn, J. A. Farmer, P. H. Jones, J. R. Schein, and A. M. Gotto Jr Influence of Low HDL on Progression of Coronary Artery Disease and Response to Fluvastatin Therapy Circulation, February 16, 1999; 99(6): 736 - 743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Lamarche, A. Tchernof, P. Mauriege, B. Cantin, G. R. Dagenais, P. J. Lupien, and J.-P. Despres Fasting Insulin and Apolipoprotein B Levels and Low-Density Lipoprotein Particle Size as Risk Factors for Ischemic Heart Disease JAMA, June 24, 1998; 279(24): 1955 - 1961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Stein and P. E. McBride Hyperhomocysteinemia and Atherosclerotic Vascular Disease: Pathophysiology, Screening, and Treatment Arch Intern Med, June 22, 1998; 158(12): 1301 - 1306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. K. Shah, J. Nilsson, S. Kaul, M. C. Fishbein, H. Ageland, A. Hamsten, J. Johansson, F. Karpe, and B. Cercek Effects of Recombinant Apolipoprotein A-IMilano on Aortic Atherosclerosis in Apolipoprotein E–Deficient Mice Circulation, March 3, 1998; 97(8): 780 - 785. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Stein and R. S. Rosenson Lipoprotein Lp(a) Excess and Coronary Heart Disease Arch Intern Med, June 9, 1997; 157(11): 1170 - 1176. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||