C-Reactive Protein—Cause or Effect?
During the past month, C-reactive protein (CRP) has been much in the news. CRP has been likened to a Cinderella, dwelling in obscurity for many years, then being magically uplifted and taken in a coach to the ball. The supporter in this instance is Paul M. Ridker, MD, from Brigham and Women’s Hospital in Boston, Mass, whose study on 27 939 women generated astonishment by showing CRP to be a stronger predictor of cardiovascular events than LDL cholesterol. One message of his report was that CRP might be a useful means of screening people who are at excess risk of vascular disease despite normal LDL cholesterol; 46% of first cardiovascular events were in women with an LDL <130 mg/dL (3.36 mmol/L). Providing a foil for Dr Ridker, the part of critical commentator was played by Lori Mosca, MD, of Columbia University in New York, who was skeptical about this proposal. Where, she asked, is the evidence that CRP improves usefully on conventional markers? Statistical significance is not clinical significance. Until better data emerge, “It is prudent to focus effort and resources on screening for and treatment of major conventional risk factors.”
This scene, played out originally in the November 14, 2002, issue of the New England Journal of Medicine (N Engl J Med. 2002;347:1557–1565, 1615–1616), was reenacted with an additional cast at the American Heart Association’s (AHA) 2002 Scientific Sessions in Chicago, Ill. Moderating the debate, Sidney C. Smith Jr, MD, from the University of North Carolina in Chapel Hill (and the AHA’s Chief Science Officer) leaned toward the view that CRP screening would be useful in primary prevention by offering an additional way to motivate people to adopt healthier lifestyles. But there were numerous matters to be resolved first, such as who should be tested, with what cut-offs, how many readings, and so on. The AHA and the Centers for Disease Control and Prevention are now struggling to produce joint guidelines. Other aspects of the protein were aired in the Chicago debate. In future drug trials, CRP could provide a useful warning of cardiovascular hazard: for instance, armed with today’s information, investigators might have been more concerned about the high CRP associated with hormone replacement therapy, whose ill-effects on coronary arteries were confirmed elsewhere during the Chicago Sessions. More intriguing is the notion that CRP has a causal role in the inflammation that is thought to underlie coronary heart disease. Edward T.H. Yeh, MD, from the University of Texas Health Science Center in Houston, discussed the inducing effects of CRP upon adhesion molecules on coronary endothelial cells, which might cause inflammatory cells to adhere and precipitate atherothrombosis. There is no reason, of course, to think CRP is the only molecule that acts in this way; there may be many. Indeed, elsewhere at the Sessions, Robert P. Palusinski, MD, also from the University of Texas Health Science Center, reported evidence of synergy between CRP and CD14, a protein that binds to bacterial endotoxin. If, however, CRP is proinflammatory, treatments that depress this protein might be protective, and the drug options include fenofibrate and statins, though not seemingly aspirin. At the meeting, Dr Ridker announced funding of a new randomized trial (JUPITER) to explore the cardiovascular protection offered by simvastatin as a CRP depressant. The drug or placebo will be given to some 15 000 individuals with raised CRP but normal LDL (<3.36 mmol/L). This is not the perfect test of CRP causation, because the statin will affect lipids, too.
Quality Improvement: Leaders Needed
A recurring theme at AHA meetings is the chasm between knowledge and medical practice. Today in the United States, only some 54% of patients receive optimal care for a given medical condition, and the mismatch assumes greater importance as treatments become more effective. At this year’s Scientific Sessions in Chicago, Ill, two special lectures addressed the gap.
First, pharmacotherapy: what are the barriers to optimal use of drugs? In his Ancel Keys Lecture, Curt Furberg, MD, from Wake Forest University in Winston-Salem, NC, identified obstacles throughout the healthcare system: patients who decline treatment, neglect to fill a prescription, or stop taking prescribed medications; providers who deviate from recommended therapy because they are uninformed, unconvinced, or unwilling to accept guidelines; payers/institutions that withhold effective products because of cost; an industry that promotes (sometimes directly to consumers) high-cost products with no advantages over existing agents; and a Food and Drug Administration that shows scant enthusiasm for comparative trials. Part of the solution is for payers to offer special rewards for healthcare of high quality, but reform demands major and concerted efforts. Who, he asked, will provide the necessary leadership?
Among the less effective measures listed by Dr Furberg were printed material, guidelines, and large meetings. His gloomy views on guidelines, at least, were somewhat offset by two presentations at the Scientific Sessions from the Duke Clinical Research Institute in Durham, NC. Eric D. Peterson, MD, showed convincingly that guidelines, if followed, are beneficial in myocardial infarction (MI). He and his colleagues had conducted a study of MI care indicators, based on AHA/American College of Cardiology recommendations, in ≈87 000 patients with MI reported to a national registry. They found huge variations in the care indicators, and centers that adhered closely to the recommendations had the lowest in-hospital mortality rates. For example, when hospitals were grouped in quartiles (n=271 for each) by percentage of “correct” care delivered, mortality was 11.9% in the leading quartile and 17.6% in the hindmost. Then T. Bruce Ferguson, MD, reported a randomized study of strategies for securing continuous quality improvement. With cooperation from the Society of Thoracic Surgeons, his group had sought to discover whether certain low-intensity interventions would speed the adoption of two process-of-care measures in coronary artery bypass grafting (CABG). These two recommended measures are the preoperative use of β-blockers and the use of internal mammary artery rather than saphenous vein grafts in patients aged >75 years. A total of 359 sites of CABG surgery were randomized to receive interventions directed at one of these objectives or to a control group. The interventions were: (1) a call to action to a physician leader at the site; (2) distribution of educational products; and (3) longitudinal nationally benchmarked site-specific feedback. Each group had >88 000 patients. The statistical analysis was necessarily complex, but the results showed clear benefit in the β-blocker group; in the internal-mammary-artery group there was a nonsignificant gain. This randomized trial, according to the Duke investigators, is a potential model for large-scale quality improvement efforts across all disciplines of medicine. The intervention was, of course, physician-led.
In his Lewis A. Conner Memorial Lecture, Kenneth A. Shine, MD, of the Rand Corporation, took a broader view than Dr Furberg. His theme was error through system failure. As president of the Institute of Medicine, he had lately delivered an alarming report on medical errors in the healthcare system—for which the death toll was higher than for breast cancer, AIDS, or road accidents. Revisiting this report, Dr Shine emphasized that the catalogue of error was not, in general, a result of poor clinical judgment; rather, most of the errors arose from failure to execute treatments as planned, or from basing treatment plans on erroneous data. Underlying these system failures, he declared, was a mode of medicine that must now be abandoned. In the 20th century, physicians had clung to their autonomy; moreover, blaming and shaming, part of medical education, was central to a tort system that increased healthcare costs, encouraged defensive medicine, and limited interactions between healthcare workers. The 21st century paradigm must be the team, in which the physician is a leader but other health professionals and patients are key players. Physicians must be linked through information systems that identify good care and the system must be transparent to all its participants. Patients should have access to their own records. Blame/shame must be replaced by mutual problem solving. “Tort reform is essential.”
In Dr Shine’s ideal world, the focus will switch from cost to value, with patients at the center of the system. He offered a partial answer to Dr Furberg’s question about leadership. In cardiovascular medicine, where better than in the AHA, which includes a full spectrum of team members—physicians, nurses, pharmacists, emergency care workers, public health specialists, nutritionists and patients.