Theodore G. Ganiats, MD

Diabetes is a very common disease, and amputations are a significant adverse outcome. Physicians feel compelled to do something to prevent these amputations. Logic dictates that sensory neuropathy predates the amputation and is causally related to the trauma that precedes it. Therefore, early detection of neuropathy may lead to interventions that decrease the likelihood of amputations. This has been the focus of a large number of studies and of current medical practice.

As Mayfield and colleagues point out in this issue of the Journal, many studies have evaluated the effectiveness of routine foot examinations for reducing amputation risk. These evaluations, however, have been part of a whole-patient approach to diabetes care. This is unfortunate, because it is not possible to evaluate the independent protective effect of foot examinations when they are part of a multitiered treatment plan. A patient who participates in such a treatment program may have reduced amputation risk, but it is not possible to tell whether the foot examinations led to this reduction or if it was caused by some other factor (such as better diabetes control).

Evaluating the evidence

There are several types of interventions in medicine. We evaluate treatment interventions by measuring outcomes in a randomized trial in which some patients are treated and some are not. Showing that there is an effective treatment is useful information, but a single trial cannot provide evidence about the intensity or duration of the treatment. For example, 30 years ago we treated uncomplicated urinary tract infections for 10 days because that was what the trials and clinical experience dictated. Later, shorter-course treatments became popular when clinical trials showed that a shorter course provided equal effectiveness and fewer side effects. The original trials showed that the treatment was effective but did not tell us the optimal treatment length.

Screening is another type of intervention. We can also evaluate the efficacy of screening by randomizing patients to screening and nonscreening groups and following the outcomes. If screening is shown to be effective, we have important information but still do not know how often it is needed. We can show that performing a Papanicolau (Pap) test every month reduces the risk of cervical cancer, but this does not prove that monthly is the optimal frequency.

Examining the foot of a person with diabetes is a screening procedure, and it would be nice to have randomized trials to evaluate its effectiveness. However, it is difficult to do these trials once the foot examination is a standard of care. In addition, although not uncommon, most patients with diabetes do not receive an amputation, and there is a long period from the onset of diabetes to an amputation. Both of these factors also argue against a clinical trial. Without a randomized trial, we are forced to approach the clinical question using other methods.

Epidemiologic methods, such as the case control study used by Mayfield and coworkers, can be useful. It is impossible, however, to prove causation through an epidemiologic study. Fortunately, there are several questions we can ask of epidemiologic data to evaluate the likelihood and the strength of a causal relationship. Is the evidence from humans? Obviously, animal studies do not provide as much support for causation as human studies. If there is an association, is the association strong? Strong associations are more likely to be tied to causal relationships than weak associations. Is the association consistent? Causal relationships are more likely if every study looking at the association shows the same findings.

We can also ask: Is there a dose-response gradient? This one is very tricky, because the doses used in the literature must be appropriate to see the gradient. You could show no gradient in the reduction of cervical cancer as you move from weekly to monthly to yearly Pap tests. This lack of gradient argues against Pap tests reducing the risk of cervical cancer. However, if the studies had evaluated Pap tests performed annually, every third year, and every sixth year, the appropriate dose response would have been seen.

Another evaluative question is: Is the temporal relationship correct? Although this is helpful with some topics (eg, low cholesterol measured before myocardial infarction), it is obvious that the foot examination must occur before the amputation.

Foot examinations

In the article by Mayfield and colleagues, we see that people who had an amputation had more foot examinations than people who did not. This suggests another difficulty in evaluating interventions: Often those who are more ill are more likely to be screened. Does the benefit documented in the study extend to all patients, or is it confined to those with advanced disease?

Mayfield and coworkers provide important additional evidence supporting the inclusion of foot examinations as a routine part of the evaluation of patients with diabetes. The article by itself does not prove that these examinations are useful. Even with the whole weight of evidence in the literature, this is still a contentious issue. However, the study by Mayfield and colleagues lends additional support to this practice.

There are several limitations to their study, however. The study used a highly select population. As the authors note, the Gila River Indian community has one of the world’s highest reported rates of diabetes and amputations. It may be that diabetes in this population is a fundamentally different disease, and that amputations are more likely and foot examinations are less effective than in the general population. If this were the case, the findings of this study grossly underestimate the true impact of foot examinations. Or it may be that in this population foot examinations are more likely to be beneficial than in another population. If that were the case, the study grossly overestimates the impact of foot examinations for reducing amputation risk.

Each year our government spends billions of dollars on research, but some basic clinical questions remain unanswered. Should we adopt foot screening as a standard in diabetes care? If not, is there a subset of patients that would benefit from screening? If we do screen, how often should it be done? Each visit? Annually? Even simple questions like "What constitutes a quality foot examination?" go unanswered.

The answers may not be determined soon, but family physicians can play a key role in finding them. Our practice-based research networks contain a large number of patients with diabetes, and carefully designed studies can assess either retrospective or even prospective evaluations of diabetes care. Diabetes care is a major challenge, and family physicians provide the bulk of this care to patients. Perhaps we will also provide the answers to some of the key questions about quality care for this disease.

Acknowledgments

Dr Ganiats is a recipient of the American Academy of Family Physicians Advanced Research Training Grant.

Diabetes is a very common disease, and amputations are a significant adverse outcome. Physicians feel compelled to do something to prevent these amputations. Logic dictates that sensory neuropathy predates the amputation and is causally related to the trauma that precedes it. Therefore, early detection of neuropathy may lead to interventions that decrease the likelihood of amputations. This has been the focus of a large number of studies and of current medical practice.

As Mayfield and colleagues point out in this issue of the Journal, many studies have evaluated the effectiveness of routine foot examinations for reducing amputation risk. These evaluations, however, have been part of a whole-patient approach to diabetes care. This is unfortunate, because it is not possible to evaluate the independent protective effect of foot examinations when they are part of a multitiered treatment plan. A patient who participates in such a treatment program may have reduced amputation risk, but it is not possible to tell whether the foot examinations led to this reduction or if it was caused by some other factor (such as better diabetes control).

Evaluating the evidence

There are several types of interventions in medicine. We evaluate treatment interventions by measuring outcomes in a randomized trial in which some patients are treated and some are not. Showing that there is an effective treatment is useful information, but a single trial cannot provide evidence about the intensity or duration of the treatment. For example, 30 years ago we treated uncomplicated urinary tract infections for 10 days because that was what the trials and clinical experience dictated. Later, shorter-course treatments became popular when clinical trials showed that a shorter course provided equal effectiveness and fewer side effects. The original trials showed that the treatment was effective but did not tell us the optimal treatment length.

Screening is another type of intervention. We can also evaluate the efficacy of screening by randomizing patients to screening and nonscreening groups and following the outcomes. If screening is shown to be effective, we have important information but still do not know how often it is needed. We can show that performing a Papanicolau (Pap) test every month reduces the risk of cervical cancer, but this does not prove that monthly is the optimal frequency.

Examining the foot of a person with diabetes is a screening procedure, and it would be nice to have randomized trials to evaluate its effectiveness. However, it is difficult to do these trials once the foot examination is a standard of care. In addition, although not uncommon, most patients with diabetes do not receive an amputation, and there is a long period from the onset of diabetes to an amputation. Both of these factors also argue against a clinical trial. Without a randomized trial, we are forced to approach the clinical question using other methods.

Epidemiologic methods, such as the case control study used by Mayfield and coworkers, can be useful. It is impossible, however, to prove causation through an epidemiologic study. Fortunately, there are several questions we can ask of epidemiologic data to evaluate the likelihood and the strength of a causal relationship. Is the evidence from humans? Obviously, animal studies do not provide as much support for causation as human studies. If there is an association, is the association strong? Strong associations are more likely to be tied to causal relationships than weak associations. Is the association consistent? Causal relationships are more likely if every study looking at the association shows the same findings.

We can also ask: Is there a dose-response gradient? This one is very tricky, because the doses used in the literature must be appropriate to see the gradient. You could show no gradient in the reduction of cervical cancer as you move from weekly to monthly to yearly Pap tests. This lack of gradient argues against Pap tests reducing the risk of cervical cancer. However, if the studies had evaluated Pap tests performed annually, every third year, and every sixth year, the appropriate dose response would have been seen.

Another evaluative question is: Is the temporal relationship correct? Although this is helpful with some topics (eg, low cholesterol measured before myocardial infarction), it is obvious that the foot examination must occur before the amputation.

Foot examinations

In the article by Mayfield and colleagues, we see that people who had an amputation had more foot examinations than people who did not. This suggests another difficulty in evaluating interventions: Often those who are more ill are more likely to be screened. Does the benefit documented in the study extend to all patients, or is it confined to those with advanced disease?

Mayfield and coworkers provide important additional evidence supporting the inclusion of foot examinations as a routine part of the evaluation of patients with diabetes. The article by itself does not prove that these examinations are useful. Even with the whole weight of evidence in the literature, this is still a contentious issue. However, the study by Mayfield and colleagues lends additional support to this practice.

There are several limitations to their study, however. The study used a highly select population. As the authors note, the Gila River Indian community has one of the world’s highest reported rates of diabetes and amputations. It may be that diabetes in this population is a fundamentally different disease, and that amputations are more likely and foot examinations are less effective than in the general population. If this were the case, the findings of this study grossly underestimate the true impact of foot examinations. Or it may be that in this population foot examinations are more likely to be beneficial than in another population. If that were the case, the study grossly overestimates the impact of foot examinations for reducing amputation risk.

Each year our government spends billions of dollars on research, but some basic clinical questions remain unanswered. Should we adopt foot screening as a standard in diabetes care? If not, is there a subset of patients that would benefit from screening? If we do screen, how often should it be done? Each visit? Annually? Even simple questions like "What constitutes a quality foot examination?" go unanswered.

The answers may not be determined soon, but family physicians can play a key role in finding them. Our practice-based research networks contain a large number of patients with diabetes, and carefully designed studies can assess either retrospective or even prospective evaluations of diabetes care. Diabetes care is a major challenge, and family physicians provide the bulk of this care to patients. Perhaps we will also provide the answers to some of the key questions about quality care for this disease.

Acknowledgments

Dr Ganiats is a recipient of the American Academy of Family Physicians Advanced Research Training Grant.

Diabetes is a very common disease, and amputations are a significant adverse outcome. Physicians feel compelled to do something to prevent these amputations. Logic dictates that sensory neuropathy predates the amputation and is causally related to the trauma that precedes it. Therefore, early detection of neuropathy may lead to interventions that decrease the likelihood of amputations. This has been the focus of a large number of studies and of current medical practice.

As Mayfield and colleagues point out in this issue of the Journal, many studies have evaluated the effectiveness of routine foot examinations for reducing amputation risk. These evaluations, however, have been part of a whole-patient approach to diabetes care. This is unfortunate, because it is not possible to evaluate the independent protective effect of foot examinations when they are part of a multitiered treatment plan. A patient who participates in such a treatment program may have reduced amputation risk, but it is not possible to tell whether the foot examinations led to this reduction or if it was caused by some other factor (such as better diabetes control).

Evaluating the evidence

There are several types of interventions in medicine. We evaluate treatment interventions by measuring outcomes in a randomized trial in which some patients are treated and some are not. Showing that there is an effective treatment is useful information, but a single trial cannot provide evidence about the intensity or duration of the treatment. For example, 30 years ago we treated uncomplicated urinary tract infections for 10 days because that was what the trials and clinical experience dictated. Later, shorter-course treatments became popular when clinical trials showed that a shorter course provided equal effectiveness and fewer side effects. The original trials showed that the treatment was effective but did not tell us the optimal treatment length.

Screening is another type of intervention. We can also evaluate the efficacy of screening by randomizing patients to screening and nonscreening groups and following the outcomes. If screening is shown to be effective, we have important information but still do not know how often it is needed. We can show that performing a Papanicolau (Pap) test every month reduces the risk of cervical cancer, but this does not prove that monthly is the optimal frequency.

Examining the foot of a person with diabetes is a screening procedure, and it would be nice to have randomized trials to evaluate its effectiveness. However, it is difficult to do these trials once the foot examination is a standard of care. In addition, although not uncommon, most patients with diabetes do not receive an amputation, and there is a long period from the onset of diabetes to an amputation. Both of these factors also argue against a clinical trial. Without a randomized trial, we are forced to approach the clinical question using other methods.

Epidemiologic methods, such as the case control study used by Mayfield and coworkers, can be useful. It is impossible, however, to prove causation through an epidemiologic study. Fortunately, there are several questions we can ask of epidemiologic data to evaluate the likelihood and the strength of a causal relationship. Is the evidence from humans? Obviously, animal studies do not provide as much support for causation as human studies. If there is an association, is the association strong? Strong associations are more likely to be tied to causal relationships than weak associations. Is the association consistent? Causal relationships are more likely if every study looking at the association shows the same findings.

We can also ask: Is there a dose-response gradient? This one is very tricky, because the doses used in the literature must be appropriate to see the gradient. You could show no gradient in the reduction of cervical cancer as you move from weekly to monthly to yearly Pap tests. This lack of gradient argues against Pap tests reducing the risk of cervical cancer. However, if the studies had evaluated Pap tests performed annually, every third year, and every sixth year, the appropriate dose response would have been seen.

Another evaluative question is: Is the temporal relationship correct? Although this is helpful with some topics (eg, low cholesterol measured before myocardial infarction), it is obvious that the foot examination must occur before the amputation.

Foot examinations

In the article by Mayfield and colleagues, we see that people who had an amputation had more foot examinations than people who did not. This suggests another difficulty in evaluating interventions: Often those who are more ill are more likely to be screened. Does the benefit documented in the study extend to all patients, or is it confined to those with advanced disease?

Mayfield and coworkers provide important additional evidence supporting the inclusion of foot examinations as a routine part of the evaluation of patients with diabetes. The article by itself does not prove that these examinations are useful. Even with the whole weight of evidence in the literature, this is still a contentious issue. However, the study by Mayfield and colleagues lends additional support to this practice.

There are several limitations to their study, however. The study used a highly select population. As the authors note, the Gila River Indian community has one of the world’s highest reported rates of diabetes and amputations. It may be that diabetes in this population is a fundamentally different disease, and that amputations are more likely and foot examinations are less effective than in the general population. If this were the case, the findings of this study grossly underestimate the true impact of foot examinations. Or it may be that in this population foot examinations are more likely to be beneficial than in another population. If that were the case, the study grossly overestimates the impact of foot examinations for reducing amputation risk.

Each year our government spends billions of dollars on research, but some basic clinical questions remain unanswered. Should we adopt foot screening as a standard in diabetes care? If not, is there a subset of patients that would benefit from screening? If we do screen, how often should it be done? Each visit? Annually? Even simple questions like "What constitutes a quality foot examination?" go unanswered.

The answers may not be determined soon, but family physicians can play a key role in finding them. Our practice-based research networks contain a large number of patients with diabetes, and carefully designed studies can assess either retrospective or even prospective evaluations of diabetes care. Diabetes care is a major challenge, and family physicians provide the bulk of this care to patients. Perhaps we will also provide the answers to some of the key questions about quality care for this disease.

Acknowledgments

Dr Ganiats is a recipient of the American Academy of Family Physicians Advanced Research Training Grant.

There is a well-described gap in clinical medicine. Research is published at an ever-increasing rate, and it is far beyond the capabilities of any individual to stay abreast of all the latest developments. Despite this rapid advance in knowledge, most published data describe advances in disease-oriented, cell-oriented, or molecule-oriented medicine. These advances are seldom applicable to clinical medicine.

This is not a small matter. The first 2 articles in a recent issue of the New England Journal of Medicine were dedicated to the prophylaxis and treatment of gastritis and ulcers induced by nonsteroidal anti-inflammatory drugs (NSAIDs).^1,2 These 2 articles were disease-oriented evidence (DOEs), emphasizing NSAID-induced ulcer management and using disease-oriented (endoscopic evidence of ulcer) rather than patient-oriented (pain or other symptoms) outcomes. The articles received much publicity, especially from the developers of the “better” treatment. This publicity, both through possible direct-to-consumer advertisements and academic detailing aimed at supporting DOE-based behavior, influences physician behavior.

Of course, all fault does not lie with the medical literature, since clinicians do not always follow patient-oriented evidence that matters (POEMs). For example, a series of POEMs stretching across more than a decade demonstrated that post-myocardial infarction (MI) patients experience improved outcomes (with such significant patient-oriented outcomes as survival) when given a b-blocker.^3-5 Still, many candidates for such treatment are not given a b-blocker after an MI.

All POEMs are important; those demonstrating clinical effectiveness, however, are of the greatest value to physicians. Clinical trials are designed to assess either efficacy or effectiveness.⁶ Efficacy describes biology (ie, whether a given intervention works under ideal conditions). Most randomized trials are efficacy studies. In family medicine this is important (we like to know that something works), but efficacy is not as significant as effectiveness. Effectiveness confirms that the intervention works in the context of real world problems, such as issues of compliance (both physicians’ and patients’) and competing priorities. Efficacy is a valuable measurement tool for the Food and Drug Administration and the National Institutes of Health. Effectiveness is the measurement of a patient’s experience.

The problem is two-fold. First, there are fewer POEMs than DOEs. Second, even when provided with the appropriate POEMs (ie, those about effectiveness), physicians are often slow to change their practices. This article addresses the first issue, the relative paucity of POEMs for guiding clinical practice. What should physicians do? Should they abandon all evidence-based science? Should they blindly rely on DOEs while waiting for POEMs? Is there another alternative?

Many physicians have practiced an alternative for years. This method includes liberal use of DOEs (when available) combined with thoughtful use of causal pathways to provide preliminary direction to guide clinical decisions. This article applies that method to an example from the growing basic science surrounding endothelial functioning. The endo-thelium was chosen because its basic science is not everyday reading material for the family physician. Thus, this topic provides an excellent model for applying these techniques to any new knowledge.

The causal pathway

A causal pathway is a description of how physicians view the pathophysiology of disease. Usually the causal pathway reflects elements that are temporally and causally related. A causal pathway may be simple (eg, hypertension leads to atherosclerosis, which leads to stroke); however, a detailed pathway is more useful. Figure 1 depicts a simplified pathway for coronary heart disease.⁷ This pathway is based on the assumption that lowering high-serum cholesterol results in a lower risk of coronary heart disease. This pathway further assumes that early detection of hypercholesterolemia is possible.

A causal pathway has several uses. First, it explicitly states our understanding of mechanisms of action. This leads to testable hypotheses that can advance our knowledge. Second, it provides a series of proposed links between an early potential causative agent and an outcome. This is important, because the proposed links provide a mechanism for testing several smaller questions instead, of a large one. This has obvious cost and time advantages. The causal pathway in Figure 1 can be used as an example. A question may arise about whether screening for hypercholesterolemia can lead to a lower risk of coronary heart disease. The best way to evaluate that question is through a randomized trial as depicted in line 5 of Figure 1. However, such a trial would take a significant period of time and be expensive. Instead a series of smaller studies, each testing a different link in the causal pathway (eg, 1, 2, and 3, or 1 and 4 in Figure 1), could provide important evidence. Certainly, the larger trial is superior. However, when faced with a lack of documented evidence, supportive evidence from the links of a causal pathway can be reassuring. This is precisely what occurred during the past 20 years to establish the link between cholesterol and coronary heart disease. First, a series of epidemiologic POEMs followed by clinical trial DOEs supported the cholesterol-cardiac disease causal pathway. Recently, randomized controlled trial POEMs provided the final support for the model. Likewise, nonsupportive evidence, while not eliminating a causal link, suggests that the cause-and-effect relationship is less likely.

Carrying the cholesterol example further, many clinicians have long believed that screening for hypercholesterolemia is justified. This is because they trusted the previously described causal pathway. Only recently has the clinical trial evidence provided us with POEMs that support cholesterol screening. Initially, POEM evidence confirmed that lowering cholesterol had beneficial effects on cardiovascular disease in those with a history of myocardial infarction.^8-9 Later data supported cholesterol lowering in those patients without a previous history of myocardial infarction.^10-11 The final link (confirming that screening the entire population is justified) is not complete, but the bulk of POEM evidence in conjunction with the causal pathway supports this hypothesis.

When faced with a lack of POEM evidence in this example, physicians had the choice of waiting for POEMs or acting on the causal pathway that was supported by logic, an understanding of pathophysiology, and DOEs. Most groups, such as the National Cholesterol Educational Program (NCEP) adopted the latter approach, the results of which have since been validated. Other causal pathways, such as those supporting mammography before age 50 years and prostate-specific antigen screening for prostate cancer, are awaiting more DOEs and POEMs to validate the proposed causal pathway links. Until then, subjects of this type are open to intense controversy.

Rationale for mechanisms

Although POEMs are helpful for developing patient care strategies, it is still necessary to understand underlying pathophysiologic and treatment mechanisms. Knowing that a low-fat diet reduces heart disease is important but not as important as knowing how it reduces heart disease. This knowledge is useful in several ways. First, understanding the mechanism facilitates the learning of new concepts. For example, understanding the Frank-Starling principle helps the clinician understand a wide range of physiologic phenomena, from the response of some murmurs to physical examination maneuvers, to predicting a response to a variety of pharmacotherapies. Second, understanding the mechanism allows researchers to develop new treatments (eg, lipid-lowering medications) and helps physicians assess which other treatments are most likely to work. To take an extreme example, we may recognize that diets high in fruits and vegetables are heart-healthy but also know that our patient population prefers fried fast food. Without realizing that one mechanism by which fruits and vegetables improve cardiovascular health is by decreasing fat intake, we may mistakenly urge our patients to consume deep-fried vegetables. Clearly, for both the researcher and the clinician, an understanding of mechanisms is helpful.

Endothelium dysfunction

A Prospective Example. Recent evidence, both POEMs and DOEs, supports the use of ACE inhibitors for the treatment of a variety of cardiovascular diseases.^12-17 Why do angiotensin-converting enzyme (ACE) inhibitors work? Is there a common pathway? Would an understanding of the mechanisms change physicians’ understanding of the causal pathway for cardiovascular disease? If so, will that affect treatment? To best address these questions it is important to review endothelial physiology.

Endothelial Physiology and Pathophysiology. The endothelium was once considered a relatively inert barrier that allowed diffusion between the blood and the vascular smooth muscle.¹⁸ It is now recognized that the endothelium is the largest internal organ. With more than a trillion cells, it has a mass greater than the liver. In a 70-kg man, the total vascular surface area is equivalent to 6 tennis courts.^18,19 More important, the endothelium is recognized as an active organ responsible for a large number of critical functions, some of which are summarized in Table 1.¹⁸ The endothelium has numerous endocrine and paracrine functions. For example, it senses hemodynamic forces and hormonal changes around the vasculature and responds by synthesizing and releasing biologically active substances (Figure 2).¹⁸ Release of these substances controls or moderates vascular tone, vascular remodeling, hemostasis and thrombosis, and inflammation. (Vascular tone is reviewed in this report. Details of the other actions are reviewed elsewhere.^20-23) Recent evidence supports the belief that the endothelium is central to the causal pathway depicted in Figure 3.¹⁸ This figure demonstrates how a variety of risk factors other than hypercholesterolemia may interact and how the entire cardiovascular disease spectrum is interrelated.

Perhaps the most critical of the endothelium’s functions is the maintenance of vascular tone. Vascular relaxation and contraction are accomplished through the production of several factors that have an impact on the underlying vascular smooth muscle. Nitric oxide (NO) is an important and potent vasodilator^24,25 and an inhibitor of platelet aggregation. It also plays a role in cardiac contractility, endothelial permeability, endothelial-leukocyte interactions, and thrombosis.²⁶ Bradykinin is another vasodilator that works both directly on the smooth muscle and indirectly by stimulating the release of NO.^27-29 Because it is also a potent stimulator of tissue plasminogen activator (tPA) secretion, bradykinin has beneficial antithrombotic effects. 25 Several substances stimulate endothelial-dependent vascular contractions. For example, acetylcholine, nicotine, and hypoxia stimulate contraction through the endothelium. The endothelium also regulates the release of the vasoconstricting agents thromboxane A2 and angiotensin II.^19,28

ACEs occurs in both a circulating and a tissue form (ie, endothelial tissue ACE). Interestingly, ACEs affect both sides of the endothelial balance by stimulating the production of angiotensin II (a vasoconstrictor) and reducing bradykinin (a vasodilator) by converting it to an inactive substance. Thus, ACEs have a significant impact on the vasculature, summarized in Table 2.¹⁸ In a state of endothelial dysfunction, an imbalance of vasoactive regulators results in higher levels of ACEs (which promote vasoconstriction, vascular remodeling, coagulation, and inflammation) and inhibits bradykinin and the release of NO (which promotes vasodilation, inhibits vascular remodeling, stimulates the release of tPA, and reduces inflammation).

Establishing the Significance of the Causal Pathway. This biochemical picture produces a nice snapshot that may have clinical relevance, but the central question for the family physician is still: So what? All of the basic scientific research and the DOEs that produce the evidence in this section offer little consolation to the clinician. Theory is nice; outcomes are critical.

Evidence of clinical relevance may be found in the fact that several factors are associated with endothelial dysfunction, which establishes an important link between endothelial function and disease. These factors include atherosclerosis, heart failure, hypertension, hypercholesterolemia, cigarette smoking, insulin resistance/diabetes mellitus, withdrawal of estrogen, and homocysteine. Thus, many major cardiovascular risk factors are associated with endothelial dysfunction (Figure 3).¹⁸ Because the endothelium regulates vasodilation, inhibition of vascular smooth muscle growth, inflammation, and antithrombotic factors, endothelial dysfunction is associated with vasoconstriction, vascular smooth muscle growth, inflammation, and thrombosis.

Fortunately, there are a large number of studies supporting this research.³⁰ Several recent clinical studies support the model represented in Figure 3.¹⁸ Some of these are POEMs, but the majority are DOEs.

CLINICAL REVIEWS DOEs

Associations are important observations, but they do not demonstrate causation. Further evidence supporting the endothelial-cardiovascular link is found in DOEs. For example, diet and lifestyle changes, such as physical exercise and smoking cessation, have been shown to improve endothelial function. The administration of antioxidants, lipid-lowering agents, estrogens (in women), calcium antagonists, and ACE inhibitors have also been shown to improve endothelial function. ACE inhibitors are especially interesting, because they offer a pharmacologic approach that can be used in conjunction with lifestyle changes.

A large number of DOEs are underway or have recently been completed that evaluate the clinical implications of the described biochemistry. For example, the Trial on Reversing Endothelial Dysfunction¹⁷ evaluated the response of endothelium-dependent vasodilation to the ACE inhibitor quinapril. This study confirmed the hypothesis that ACE inhibitors improve endothelial function in patients with documented endothelial dysfunction. Other studies, soon be completed, will add to the body of disease-oriented knowledge supporting the central role of the endothelium in cardiovascular disease.

POEMs

Only one major POEM-based study has been conducted in this field. The investigators of the Quinapril Ischemic Events Trial evaluated approximately 1700 patients with a recent percutaneous transluminal coronary angioplasty. Unpublished results indicate that cardiac ischemic end points (cardiovascular death, nonfatal MI, need for revascularization, unstable angina) were improved with the use of an ACE inhibitor. It is interesting to note that the primary outcome for all of these studies is either death or cardiovascular morbidity. Work on other POEM-relevant outcomes, such as other morbidities, side effects, and patient preferences, has not been completed. In addition, studies in other patient groups that are more relevant to primary care, such as those focused on primary or secondary prevention, are needed.

Discussion

Limitations of Causal Pathways

Because of a lack of POEMs—notably POEM evidence of effectiveness—physicians may feel uncertain about proper treatment and fear being misguided. Causal pathways offer important assistance because they provide the logic to help physicians fill in the gaps while waiting for the POEMs to arrive. Finding evidence for each link in the causal pathway provides the support physicians need. However, causal pathways can deceive. As an example, there was an excellent (but wrong) causal pathway that suggested internal mammary artery ligation would benefit coronary artery disease patients.³¹ The proposed mechanism was that ligation of the artery would force more blood to the myocardium. Clinical experience supported the procedure, since many patients noted fewer symptoms.³¹ However, a randomized controlled trial including a sham surgery clarified the misunderstanding and eliminated the pathway.³² More recently, oncologists are evaluating causal pathways suggesting that early diagnosis of breast cancer in those younger than 50 years^32,33 or of prostate cancer³⁵ is beneficial. However, early detection of lung cancer and ovarian cancer offers little benefit.^35-37 Causal pathways must be based on hard evidence, but still carefully evaluated using common sense and experience.

The bottom line

What do we do until the POEMs arrive? Clearly, we can wait for POEMs, but such delays would paralyze clinicians and deny patients many excellent treatments. POEMs, especially POEMs of effectiveness, should help mold practices. Practicing without POEMs has risks, so it is necessary to carefully evaluate the existing evidence in light of the potential benefits, the potential harms, and the likelihood of these outcomes.

For the example of ACE inhibitors, the case seems fairly clear-cut. The causal pathway fits a growing basic science. Extensive DOEs and POEMs have provided support for the links of the causal pathway, reassuring physicians that the pathway is relevant to practice. In addition, ACE inhibitors show a growing usefulness for noncardiovascular diseases, such as diabetic nephropathy, and for nonvascular cardiac disease, such as congestive heart failure. As a class, ACE inhibitors are generally well tolerated in most patients.^17,38-40 Until the final POEMs arrive, the use of ACE inhibitors in a variety of cardiovascular diseases seems justified.

When faced with a sufficient weight of POEMs, patients can be shown good evidence to support suggested treatment strategies. Without this evidence, it is necessary to weigh the pros and cons, including the likelihood of benefit and harm, and make a decision. An understanding of pathophysiologic mechanisms helps physicians construct good causal pathways. Finding evidence, usually in the form of DOEs, to support the links of the pathway is often the best way to proceed while waiting for the POEMs to arrive.

Acknowledgments

This work was supported, in part, by a grant from Medical Education Systems, Inc.

There is a well-described gap in clinical medicine. Research is published at an ever-increasing rate, and it is far beyond the capabilities of any individual to stay abreast of all the latest developments. Despite this rapid advance in knowledge, most published data describe advances in disease-oriented, cell-oriented, or molecule-oriented medicine. These advances are seldom applicable to clinical medicine.

This is not a small matter. The first 2 articles in a recent issue of the New England Journal of Medicine were dedicated to the prophylaxis and treatment of gastritis and ulcers induced by nonsteroidal anti-inflammatory drugs (NSAIDs).^1,2 These 2 articles were disease-oriented evidence (DOEs), emphasizing NSAID-induced ulcer management and using disease-oriented (endoscopic evidence of ulcer) rather than patient-oriented (pain or other symptoms) outcomes. The articles received much publicity, especially from the developers of the “better” treatment. This publicity, both through possible direct-to-consumer advertisements and academic detailing aimed at supporting DOE-based behavior, influences physician behavior.

Of course, all fault does not lie with the medical literature, since clinicians do not always follow patient-oriented evidence that matters (POEMs). For example, a series of POEMs stretching across more than a decade demonstrated that post-myocardial infarction (MI) patients experience improved outcomes (with such significant patient-oriented outcomes as survival) when given a b-blocker.^3-5 Still, many candidates for such treatment are not given a b-blocker after an MI.

All POEMs are important; those demonstrating clinical effectiveness, however, are of the greatest value to physicians. Clinical trials are designed to assess either efficacy or effectiveness.⁶ Efficacy describes biology (ie, whether a given intervention works under ideal conditions). Most randomized trials are efficacy studies. In family medicine this is important (we like to know that something works), but efficacy is not as significant as effectiveness. Effectiveness confirms that the intervention works in the context of real world problems, such as issues of compliance (both physicians’ and patients’) and competing priorities. Efficacy is a valuable measurement tool for the Food and Drug Administration and the National Institutes of Health. Effectiveness is the measurement of a patient’s experience.

The problem is two-fold. First, there are fewer POEMs than DOEs. Second, even when provided with the appropriate POEMs (ie, those about effectiveness), physicians are often slow to change their practices. This article addresses the first issue, the relative paucity of POEMs for guiding clinical practice. What should physicians do? Should they abandon all evidence-based science? Should they blindly rely on DOEs while waiting for POEMs? Is there another alternative?

Many physicians have practiced an alternative for years. This method includes liberal use of DOEs (when available) combined with thoughtful use of causal pathways to provide preliminary direction to guide clinical decisions. This article applies that method to an example from the growing basic science surrounding endothelial functioning. The endo-thelium was chosen because its basic science is not everyday reading material for the family physician. Thus, this topic provides an excellent model for applying these techniques to any new knowledge.

The causal pathway

A causal pathway is a description of how physicians view the pathophysiology of disease. Usually the causal pathway reflects elements that are temporally and causally related. A causal pathway may be simple (eg, hypertension leads to atherosclerosis, which leads to stroke); however, a detailed pathway is more useful. Figure 1 depicts a simplified pathway for coronary heart disease.⁷ This pathway is based on the assumption that lowering high-serum cholesterol results in a lower risk of coronary heart disease. This pathway further assumes that early detection of hypercholesterolemia is possible.

A causal pathway has several uses. First, it explicitly states our understanding of mechanisms of action. This leads to testable hypotheses that can advance our knowledge. Second, it provides a series of proposed links between an early potential causative agent and an outcome. This is important, because the proposed links provide a mechanism for testing several smaller questions instead, of a large one. This has obvious cost and time advantages. The causal pathway in Figure 1 can be used as an example. A question may arise about whether screening for hypercholesterolemia can lead to a lower risk of coronary heart disease. The best way to evaluate that question is through a randomized trial as depicted in line 5 of Figure 1. However, such a trial would take a significant period of time and be expensive. Instead a series of smaller studies, each testing a different link in the causal pathway (eg, 1, 2, and 3, or 1 and 4 in Figure 1), could provide important evidence. Certainly, the larger trial is superior. However, when faced with a lack of documented evidence, supportive evidence from the links of a causal pathway can be reassuring. This is precisely what occurred during the past 20 years to establish the link between cholesterol and coronary heart disease. First, a series of epidemiologic POEMs followed by clinical trial DOEs supported the cholesterol-cardiac disease causal pathway. Recently, randomized controlled trial POEMs provided the final support for the model. Likewise, nonsupportive evidence, while not eliminating a causal link, suggests that the cause-and-effect relationship is less likely.

Carrying the cholesterol example further, many clinicians have long believed that screening for hypercholesterolemia is justified. This is because they trusted the previously described causal pathway. Only recently has the clinical trial evidence provided us with POEMs that support cholesterol screening. Initially, POEM evidence confirmed that lowering cholesterol had beneficial effects on cardiovascular disease in those with a history of myocardial infarction.^8-9 Later data supported cholesterol lowering in those patients without a previous history of myocardial infarction.^10-11 The final link (confirming that screening the entire population is justified) is not complete, but the bulk of POEM evidence in conjunction with the causal pathway supports this hypothesis.

When faced with a lack of POEM evidence in this example, physicians had the choice of waiting for POEMs or acting on the causal pathway that was supported by logic, an understanding of pathophysiology, and DOEs. Most groups, such as the National Cholesterol Educational Program (NCEP) adopted the latter approach, the results of which have since been validated. Other causal pathways, such as those supporting mammography before age 50 years and prostate-specific antigen screening for prostate cancer, are awaiting more DOEs and POEMs to validate the proposed causal pathway links. Until then, subjects of this type are open to intense controversy.

Rationale for mechanisms

Although POEMs are helpful for developing patient care strategies, it is still necessary to understand underlying pathophysiologic and treatment mechanisms. Knowing that a low-fat diet reduces heart disease is important but not as important as knowing how it reduces heart disease. This knowledge is useful in several ways. First, understanding the mechanism facilitates the learning of new concepts. For example, understanding the Frank-Starling principle helps the clinician understand a wide range of physiologic phenomena, from the response of some murmurs to physical examination maneuvers, to predicting a response to a variety of pharmacotherapies. Second, understanding the mechanism allows researchers to develop new treatments (eg, lipid-lowering medications) and helps physicians assess which other treatments are most likely to work. To take an extreme example, we may recognize that diets high in fruits and vegetables are heart-healthy but also know that our patient population prefers fried fast food. Without realizing that one mechanism by which fruits and vegetables improve cardiovascular health is by decreasing fat intake, we may mistakenly urge our patients to consume deep-fried vegetables. Clearly, for both the researcher and the clinician, an understanding of mechanisms is helpful.

Endothelium dysfunction

A Prospective Example. Recent evidence, both POEMs and DOEs, supports the use of ACE inhibitors for the treatment of a variety of cardiovascular diseases.^12-17 Why do angiotensin-converting enzyme (ACE) inhibitors work? Is there a common pathway? Would an understanding of the mechanisms change physicians’ understanding of the causal pathway for cardiovascular disease? If so, will that affect treatment? To best address these questions it is important to review endothelial physiology.

Endothelial Physiology and Pathophysiology. The endothelium was once considered a relatively inert barrier that allowed diffusion between the blood and the vascular smooth muscle.¹⁸ It is now recognized that the endothelium is the largest internal organ. With more than a trillion cells, it has a mass greater than the liver. In a 70-kg man, the total vascular surface area is equivalent to 6 tennis courts.^18,19 More important, the endothelium is recognized as an active organ responsible for a large number of critical functions, some of which are summarized in Table 1.¹⁸ The endothelium has numerous endocrine and paracrine functions. For example, it senses hemodynamic forces and hormonal changes around the vasculature and responds by synthesizing and releasing biologically active substances (Figure 2).¹⁸ Release of these substances controls or moderates vascular tone, vascular remodeling, hemostasis and thrombosis, and inflammation. (Vascular tone is reviewed in this report. Details of the other actions are reviewed elsewhere.^20-23) Recent evidence supports the belief that the endothelium is central to the causal pathway depicted in Figure 3.¹⁸ This figure demonstrates how a variety of risk factors other than hypercholesterolemia may interact and how the entire cardiovascular disease spectrum is interrelated.

Perhaps the most critical of the endothelium’s functions is the maintenance of vascular tone. Vascular relaxation and contraction are accomplished through the production of several factors that have an impact on the underlying vascular smooth muscle. Nitric oxide (NO) is an important and potent vasodilator^24,25 and an inhibitor of platelet aggregation. It also plays a role in cardiac contractility, endothelial permeability, endothelial-leukocyte interactions, and thrombosis.²⁶ Bradykinin is another vasodilator that works both directly on the smooth muscle and indirectly by stimulating the release of NO.^27-29 Because it is also a potent stimulator of tissue plasminogen activator (tPA) secretion, bradykinin has beneficial antithrombotic effects. 25 Several substances stimulate endothelial-dependent vascular contractions. For example, acetylcholine, nicotine, and hypoxia stimulate contraction through the endothelium. The endothelium also regulates the release of the vasoconstricting agents thromboxane A2 and angiotensin II.^19,28

ACEs occurs in both a circulating and a tissue form (ie, endothelial tissue ACE). Interestingly, ACEs affect both sides of the endothelial balance by stimulating the production of angiotensin II (a vasoconstrictor) and reducing bradykinin (a vasodilator) by converting it to an inactive substance. Thus, ACEs have a significant impact on the vasculature, summarized in Table 2.¹⁸ In a state of endothelial dysfunction, an imbalance of vasoactive regulators results in higher levels of ACEs (which promote vasoconstriction, vascular remodeling, coagulation, and inflammation) and inhibits bradykinin and the release of NO (which promotes vasodilation, inhibits vascular remodeling, stimulates the release of tPA, and reduces inflammation).

Establishing the Significance of the Causal Pathway. This biochemical picture produces a nice snapshot that may have clinical relevance, but the central question for the family physician is still: So what? All of the basic scientific research and the DOEs that produce the evidence in this section offer little consolation to the clinician. Theory is nice; outcomes are critical.

Evidence of clinical relevance may be found in the fact that several factors are associated with endothelial dysfunction, which establishes an important link between endothelial function and disease. These factors include atherosclerosis, heart failure, hypertension, hypercholesterolemia, cigarette smoking, insulin resistance/diabetes mellitus, withdrawal of estrogen, and homocysteine. Thus, many major cardiovascular risk factors are associated with endothelial dysfunction (Figure 3).¹⁸ Because the endothelium regulates vasodilation, inhibition of vascular smooth muscle growth, inflammation, and antithrombotic factors, endothelial dysfunction is associated with vasoconstriction, vascular smooth muscle growth, inflammation, and thrombosis.

Fortunately, there are a large number of studies supporting this research.³⁰ Several recent clinical studies support the model represented in Figure 3.¹⁸ Some of these are POEMs, but the majority are DOEs.

CLINICAL REVIEWS DOEs

Associations are important observations, but they do not demonstrate causation. Further evidence supporting the endothelial-cardiovascular link is found in DOEs. For example, diet and lifestyle changes, such as physical exercise and smoking cessation, have been shown to improve endothelial function. The administration of antioxidants, lipid-lowering agents, estrogens (in women), calcium antagonists, and ACE inhibitors have also been shown to improve endothelial function. ACE inhibitors are especially interesting, because they offer a pharmacologic approach that can be used in conjunction with lifestyle changes.

A large number of DOEs are underway or have recently been completed that evaluate the clinical implications of the described biochemistry. For example, the Trial on Reversing Endothelial Dysfunction¹⁷ evaluated the response of endothelium-dependent vasodilation to the ACE inhibitor quinapril. This study confirmed the hypothesis that ACE inhibitors improve endothelial function in patients with documented endothelial dysfunction. Other studies, soon be completed, will add to the body of disease-oriented knowledge supporting the central role of the endothelium in cardiovascular disease.

POEMs

Only one major POEM-based study has been conducted in this field. The investigators of the Quinapril Ischemic Events Trial evaluated approximately 1700 patients with a recent percutaneous transluminal coronary angioplasty. Unpublished results indicate that cardiac ischemic end points (cardiovascular death, nonfatal MI, need for revascularization, unstable angina) were improved with the use of an ACE inhibitor. It is interesting to note that the primary outcome for all of these studies is either death or cardiovascular morbidity. Work on other POEM-relevant outcomes, such as other morbidities, side effects, and patient preferences, has not been completed. In addition, studies in other patient groups that are more relevant to primary care, such as those focused on primary or secondary prevention, are needed.

Discussion

Limitations of Causal Pathways

Because of a lack of POEMs—notably POEM evidence of effectiveness—physicians may feel uncertain about proper treatment and fear being misguided. Causal pathways offer important assistance because they provide the logic to help physicians fill in the gaps while waiting for the POEMs to arrive. Finding evidence for each link in the causal pathway provides the support physicians need. However, causal pathways can deceive. As an example, there was an excellent (but wrong) causal pathway that suggested internal mammary artery ligation would benefit coronary artery disease patients.³¹ The proposed mechanism was that ligation of the artery would force more blood to the myocardium. Clinical experience supported the procedure, since many patients noted fewer symptoms.³¹ However, a randomized controlled trial including a sham surgery clarified the misunderstanding and eliminated the pathway.³² More recently, oncologists are evaluating causal pathways suggesting that early diagnosis of breast cancer in those younger than 50 years^32,33 or of prostate cancer³⁵ is beneficial. However, early detection of lung cancer and ovarian cancer offers little benefit.^35-37 Causal pathways must be based on hard evidence, but still carefully evaluated using common sense and experience.

The bottom line

What do we do until the POEMs arrive? Clearly, we can wait for POEMs, but such delays would paralyze clinicians and deny patients many excellent treatments. POEMs, especially POEMs of effectiveness, should help mold practices. Practicing without POEMs has risks, so it is necessary to carefully evaluate the existing evidence in light of the potential benefits, the potential harms, and the likelihood of these outcomes.

For the example of ACE inhibitors, the case seems fairly clear-cut. The causal pathway fits a growing basic science. Extensive DOEs and POEMs have provided support for the links of the causal pathway, reassuring physicians that the pathway is relevant to practice. In addition, ACE inhibitors show a growing usefulness for noncardiovascular diseases, such as diabetic nephropathy, and for nonvascular cardiac disease, such as congestive heart failure. As a class, ACE inhibitors are generally well tolerated in most patients.^17,38-40 Until the final POEMs arrive, the use of ACE inhibitors in a variety of cardiovascular diseases seems justified.

When faced with a sufficient weight of POEMs, patients can be shown good evidence to support suggested treatment strategies. Without this evidence, it is necessary to weigh the pros and cons, including the likelihood of benefit and harm, and make a decision. An understanding of pathophysiologic mechanisms helps physicians construct good causal pathways. Finding evidence, usually in the form of DOEs, to support the links of the pathway is often the best way to proceed while waiting for the POEMs to arrive.

Acknowledgments

This work was supported, in part, by a grant from Medical Education Systems, Inc.

There is a well-described gap in clinical medicine. Research is published at an ever-increasing rate, and it is far beyond the capabilities of any individual to stay abreast of all the latest developments. Despite this rapid advance in knowledge, most published data describe advances in disease-oriented, cell-oriented, or molecule-oriented medicine. These advances are seldom applicable to clinical medicine.

This is not a small matter. The first 2 articles in a recent issue of the New England Journal of Medicine were dedicated to the prophylaxis and treatment of gastritis and ulcers induced by nonsteroidal anti-inflammatory drugs (NSAIDs).^1,2 These 2 articles were disease-oriented evidence (DOEs), emphasizing NSAID-induced ulcer management and using disease-oriented (endoscopic evidence of ulcer) rather than patient-oriented (pain or other symptoms) outcomes. The articles received much publicity, especially from the developers of the “better” treatment. This publicity, both through possible direct-to-consumer advertisements and academic detailing aimed at supporting DOE-based behavior, influences physician behavior.

Of course, all fault does not lie with the medical literature, since clinicians do not always follow patient-oriented evidence that matters (POEMs). For example, a series of POEMs stretching across more than a decade demonstrated that post-myocardial infarction (MI) patients experience improved outcomes (with such significant patient-oriented outcomes as survival) when given a b-blocker.^3-5 Still, many candidates for such treatment are not given a b-blocker after an MI.

All POEMs are important; those demonstrating clinical effectiveness, however, are of the greatest value to physicians. Clinical trials are designed to assess either efficacy or effectiveness.⁶ Efficacy describes biology (ie, whether a given intervention works under ideal conditions). Most randomized trials are efficacy studies. In family medicine this is important (we like to know that something works), but efficacy is not as significant as effectiveness. Effectiveness confirms that the intervention works in the context of real world problems, such as issues of compliance (both physicians’ and patients’) and competing priorities. Efficacy is a valuable measurement tool for the Food and Drug Administration and the National Institutes of Health. Effectiveness is the measurement of a patient’s experience.

The problem is two-fold. First, there are fewer POEMs than DOEs. Second, even when provided with the appropriate POEMs (ie, those about effectiveness), physicians are often slow to change their practices. This article addresses the first issue, the relative paucity of POEMs for guiding clinical practice. What should physicians do? Should they abandon all evidence-based science? Should they blindly rely on DOEs while waiting for POEMs? Is there another alternative?

Many physicians have practiced an alternative for years. This method includes liberal use of DOEs (when available) combined with thoughtful use of causal pathways to provide preliminary direction to guide clinical decisions. This article applies that method to an example from the growing basic science surrounding endothelial functioning. The endo-thelium was chosen because its basic science is not everyday reading material for the family physician. Thus, this topic provides an excellent model for applying these techniques to any new knowledge.

The causal pathway

A causal pathway is a description of how physicians view the pathophysiology of disease. Usually the causal pathway reflects elements that are temporally and causally related. A causal pathway may be simple (eg, hypertension leads to atherosclerosis, which leads to stroke); however, a detailed pathway is more useful. Figure 1 depicts a simplified pathway for coronary heart disease.⁷ This pathway is based on the assumption that lowering high-serum cholesterol results in a lower risk of coronary heart disease. This pathway further assumes that early detection of hypercholesterolemia is possible.

A causal pathway has several uses. First, it explicitly states our understanding of mechanisms of action. This leads to testable hypotheses that can advance our knowledge. Second, it provides a series of proposed links between an early potential causative agent and an outcome. This is important, because the proposed links provide a mechanism for testing several smaller questions instead, of a large one. This has obvious cost and time advantages. The causal pathway in Figure 1 can be used as an example. A question may arise about whether screening for hypercholesterolemia can lead to a lower risk of coronary heart disease. The best way to evaluate that question is through a randomized trial as depicted in line 5 of Figure 1. However, such a trial would take a significant period of time and be expensive. Instead a series of smaller studies, each testing a different link in the causal pathway (eg, 1, 2, and 3, or 1 and 4 in Figure 1), could provide important evidence. Certainly, the larger trial is superior. However, when faced with a lack of documented evidence, supportive evidence from the links of a causal pathway can be reassuring. This is precisely what occurred during the past 20 years to establish the link between cholesterol and coronary heart disease. First, a series of epidemiologic POEMs followed by clinical trial DOEs supported the cholesterol-cardiac disease causal pathway. Recently, randomized controlled trial POEMs provided the final support for the model. Likewise, nonsupportive evidence, while not eliminating a causal link, suggests that the cause-and-effect relationship is less likely.

Carrying the cholesterol example further, many clinicians have long believed that screening for hypercholesterolemia is justified. This is because they trusted the previously described causal pathway. Only recently has the clinical trial evidence provided us with POEMs that support cholesterol screening. Initially, POEM evidence confirmed that lowering cholesterol had beneficial effects on cardiovascular disease in those with a history of myocardial infarction.^8-9 Later data supported cholesterol lowering in those patients without a previous history of myocardial infarction.^10-11 The final link (confirming that screening the entire population is justified) is not complete, but the bulk of POEM evidence in conjunction with the causal pathway supports this hypothesis.

When faced with a lack of POEM evidence in this example, physicians had the choice of waiting for POEMs or acting on the causal pathway that was supported by logic, an understanding of pathophysiology, and DOEs. Most groups, such as the National Cholesterol Educational Program (NCEP) adopted the latter approach, the results of which have since been validated. Other causal pathways, such as those supporting mammography before age 50 years and prostate-specific antigen screening for prostate cancer, are awaiting more DOEs and POEMs to validate the proposed causal pathway links. Until then, subjects of this type are open to intense controversy.

Rationale for mechanisms

Although POEMs are helpful for developing patient care strategies, it is still necessary to understand underlying pathophysiologic and treatment mechanisms. Knowing that a low-fat diet reduces heart disease is important but not as important as knowing how it reduces heart disease. This knowledge is useful in several ways. First, understanding the mechanism facilitates the learning of new concepts. For example, understanding the Frank-Starling principle helps the clinician understand a wide range of physiologic phenomena, from the response of some murmurs to physical examination maneuvers, to predicting a response to a variety of pharmacotherapies. Second, understanding the mechanism allows researchers to develop new treatments (eg, lipid-lowering medications) and helps physicians assess which other treatments are most likely to work. To take an extreme example, we may recognize that diets high in fruits and vegetables are heart-healthy but also know that our patient population prefers fried fast food. Without realizing that one mechanism by which fruits and vegetables improve cardiovascular health is by decreasing fat intake, we may mistakenly urge our patients to consume deep-fried vegetables. Clearly, for both the researcher and the clinician, an understanding of mechanisms is helpful.

Endothelium dysfunction

A Prospective Example. Recent evidence, both POEMs and DOEs, supports the use of ACE inhibitors for the treatment of a variety of cardiovascular diseases.^12-17 Why do angiotensin-converting enzyme (ACE) inhibitors work? Is there a common pathway? Would an understanding of the mechanisms change physicians’ understanding of the causal pathway for cardiovascular disease? If so, will that affect treatment? To best address these questions it is important to review endothelial physiology.

Endothelial Physiology and Pathophysiology. The endothelium was once considered a relatively inert barrier that allowed diffusion between the blood and the vascular smooth muscle.¹⁸ It is now recognized that the endothelium is the largest internal organ. With more than a trillion cells, it has a mass greater than the liver. In a 70-kg man, the total vascular surface area is equivalent to 6 tennis courts.^18,19 More important, the endothelium is recognized as an active organ responsible for a large number of critical functions, some of which are summarized in Table 1.¹⁸ The endothelium has numerous endocrine and paracrine functions. For example, it senses hemodynamic forces and hormonal changes around the vasculature and responds by synthesizing and releasing biologically active substances (Figure 2).¹⁸ Release of these substances controls or moderates vascular tone, vascular remodeling, hemostasis and thrombosis, and inflammation. (Vascular tone is reviewed in this report. Details of the other actions are reviewed elsewhere.^20-23) Recent evidence supports the belief that the endothelium is central to the causal pathway depicted in Figure 3.¹⁸ This figure demonstrates how a variety of risk factors other than hypercholesterolemia may interact and how the entire cardiovascular disease spectrum is interrelated.

Perhaps the most critical of the endothelium’s functions is the maintenance of vascular tone. Vascular relaxation and contraction are accomplished through the production of several factors that have an impact on the underlying vascular smooth muscle. Nitric oxide (NO) is an important and potent vasodilator^24,25 and an inhibitor of platelet aggregation. It also plays a role in cardiac contractility, endothelial permeability, endothelial-leukocyte interactions, and thrombosis.²⁶ Bradykinin is another vasodilator that works both directly on the smooth muscle and indirectly by stimulating the release of NO.^27-29 Because it is also a potent stimulator of tissue plasminogen activator (tPA) secretion, bradykinin has beneficial antithrombotic effects. 25 Several substances stimulate endothelial-dependent vascular contractions. For example, acetylcholine, nicotine, and hypoxia stimulate contraction through the endothelium. The endothelium also regulates the release of the vasoconstricting agents thromboxane A2 and angiotensin II.^19,28

ACEs occurs in both a circulating and a tissue form (ie, endothelial tissue ACE). Interestingly, ACEs affect both sides of the endothelial balance by stimulating the production of angiotensin II (a vasoconstrictor) and reducing bradykinin (a vasodilator) by converting it to an inactive substance. Thus, ACEs have a significant impact on the vasculature, summarized in Table 2.¹⁸ In a state of endothelial dysfunction, an imbalance of vasoactive regulators results in higher levels of ACEs (which promote vasoconstriction, vascular remodeling, coagulation, and inflammation) and inhibits bradykinin and the release of NO (which promotes vasodilation, inhibits vascular remodeling, stimulates the release of tPA, and reduces inflammation).

Establishing the Significance of the Causal Pathway. This biochemical picture produces a nice snapshot that may have clinical relevance, but the central question for the family physician is still: So what? All of the basic scientific research and the DOEs that produce the evidence in this section offer little consolation to the clinician. Theory is nice; outcomes are critical.

Evidence of clinical relevance may be found in the fact that several factors are associated with endothelial dysfunction, which establishes an important link between endothelial function and disease. These factors include atherosclerosis, heart failure, hypertension, hypercholesterolemia, cigarette smoking, insulin resistance/diabetes mellitus, withdrawal of estrogen, and homocysteine. Thus, many major cardiovascular risk factors are associated with endothelial dysfunction (Figure 3).¹⁸ Because the endothelium regulates vasodilation, inhibition of vascular smooth muscle growth, inflammation, and antithrombotic factors, endothelial dysfunction is associated with vasoconstriction, vascular smooth muscle growth, inflammation, and thrombosis.

Fortunately, there are a large number of studies supporting this research.³⁰ Several recent clinical studies support the model represented in Figure 3.¹⁸ Some of these are POEMs, but the majority are DOEs.

CLINICAL REVIEWS DOEs

Associations are important observations, but they do not demonstrate causation. Further evidence supporting the endothelial-cardiovascular link is found in DOEs. For example, diet and lifestyle changes, such as physical exercise and smoking cessation, have been shown to improve endothelial function. The administration of antioxidants, lipid-lowering agents, estrogens (in women), calcium antagonists, and ACE inhibitors have also been shown to improve endothelial function. ACE inhibitors are especially interesting, because they offer a pharmacologic approach that can be used in conjunction with lifestyle changes.

A large number of DOEs are underway or have recently been completed that evaluate the clinical implications of the described biochemistry. For example, the Trial on Reversing Endothelial Dysfunction¹⁷ evaluated the response of endothelium-dependent vasodilation to the ACE inhibitor quinapril. This study confirmed the hypothesis that ACE inhibitors improve endothelial function in patients with documented endothelial dysfunction. Other studies, soon be completed, will add to the body of disease-oriented knowledge supporting the central role of the endothelium in cardiovascular disease.

POEMs

Only one major POEM-based study has been conducted in this field. The investigators of the Quinapril Ischemic Events Trial evaluated approximately 1700 patients with a recent percutaneous transluminal coronary angioplasty. Unpublished results indicate that cardiac ischemic end points (cardiovascular death, nonfatal MI, need for revascularization, unstable angina) were improved with the use of an ACE inhibitor. It is interesting to note that the primary outcome for all of these studies is either death or cardiovascular morbidity. Work on other POEM-relevant outcomes, such as other morbidities, side effects, and patient preferences, has not been completed. In addition, studies in other patient groups that are more relevant to primary care, such as those focused on primary or secondary prevention, are needed.

Discussion

Limitations of Causal Pathways

Because of a lack of POEMs—notably POEM evidence of effectiveness—physicians may feel uncertain about proper treatment and fear being misguided. Causal pathways offer important assistance because they provide the logic to help physicians fill in the gaps while waiting for the POEMs to arrive. Finding evidence for each link in the causal pathway provides the support physicians need. However, causal pathways can deceive. As an example, there was an excellent (but wrong) causal pathway that suggested internal mammary artery ligation would benefit coronary artery disease patients.³¹ The proposed mechanism was that ligation of the artery would force more blood to the myocardium. Clinical experience supported the procedure, since many patients noted fewer symptoms.³¹ However, a randomized controlled trial including a sham surgery clarified the misunderstanding and eliminated the pathway.³² More recently, oncologists are evaluating causal pathways suggesting that early diagnosis of breast cancer in those younger than 50 years^32,33 or of prostate cancer³⁵ is beneficial. However, early detection of lung cancer and ovarian cancer offers little benefit.^35-37 Causal pathways must be based on hard evidence, but still carefully evaluated using common sense and experience.

The bottom line

What do we do until the POEMs arrive? Clearly, we can wait for POEMs, but such delays would paralyze clinicians and deny patients many excellent treatments. POEMs, especially POEMs of effectiveness, should help mold practices. Practicing without POEMs has risks, so it is necessary to carefully evaluate the existing evidence in light of the potential benefits, the potential harms, and the likelihood of these outcomes.

For the example of ACE inhibitors, the case seems fairly clear-cut. The causal pathway fits a growing basic science. Extensive DOEs and POEMs have provided support for the links of the causal pathway, reassuring physicians that the pathway is relevant to practice. In addition, ACE inhibitors show a growing usefulness for noncardiovascular diseases, such as diabetic nephropathy, and for nonvascular cardiac disease, such as congestive heart failure. As a class, ACE inhibitors are generally well tolerated in most patients.^17,38-40 Until the final POEMs arrive, the use of ACE inhibitors in a variety of cardiovascular diseases seems justified.

When faced with a sufficient weight of POEMs, patients can be shown good evidence to support suggested treatment strategies. Without this evidence, it is necessary to weigh the pros and cons, including the likelihood of benefit and harm, and make a decision. An understanding of pathophysiologic mechanisms helps physicians construct good causal pathways. Finding evidence, usually in the form of DOEs, to support the links of the pathway is often the best way to proceed while waiting for the POEMs to arrive.

Acknowledgments

This work was supported, in part, by a grant from Medical Education Systems, Inc.