The scientific community is constantly conducting research and clinical trials to increase the knowledge regarding the manifestation, treatment and prevention of cardiovascular disease. As new findings emerge, it becomes appropriate to update the recommendations and guidelines for the treatment and prevention of disease.
These new guidelines from the American Heart Association (AHA) and American College of Cardiology Foundation (ACCF) and endorsed by the World Heart Federation and the Preventive Cardiovascular Nurses Association (PCNA) show that control of risk factors and optimal medical treatment can help those who have experienced a heart attack or stroke, as well as those at high risk of heart attack or stroke, avoid a life threatening cardiovascular event.
SIDNEY C SMITH: It's a real pleasure to be here. And I did have the opportunity, as Doug mentioned, to come back with my son when he was at Wash U as an undergraduate and became aware of what a nice place this Saint Louis area is, and especially Washington University, Forest Park. So it's great for me to be here. What I'm going to do is sort of a gun-and-camera through the development of the new guidelines in cholesterol and hypertension, but more than just a recitation of what they say, and try to give you a feeling for what happened and what might be going on. Because both of these guidelines have turned out to be very disruptive. And disruptive behavior is somewhat in, unless you have young children, these days. One of the driving forces behind what has happened in medicine occurred at the turn of the century-- 2000-- when the Institute of Medicine came out with a statement-- "In its current form, habits, environment, American health care is incapable of providing the public with the quality of health care it expects and deserves." They went on to say that we make too many of our decisions based upon, well, that's the way I always did it, or, that's what I learned to do at Barnes or the Brigham, or whatever, and really never were able to move forward and, in a dynamic manner, incorporate new evidence into what we were doing. I didn't like this. In fact, it made me angry. I thought we were doing a pretty good job working hard. But nevertheless, the Institute of Medicine had delivered this. So I was invited shortly after that by Rob Califf, who's a good friend, to go over to Duke and speak on the guidelines. At that point, I was chairing the guidelines for ACC/AHA. And with this Institute of Medicine statement in mind just sort of on the back of a napkin, I put together the evidence based in the last five guidelines that we had done. And it turned out that only about 10% of the recommendations had strong evidence-- many RCTs. And almost 50% of them were expert opinion. And it was April. I could safely go from UNC to the Duke campus. Basketball season was over. And so I suggested that maybe we should work together. And Rob, being a good friend, agreed with that. And so we put together this paper and published it in JAMA. It was probably the most compulsive thing I'll do in my life. We looked at over 7,000 recommendations and 53 guidelines, 22 topics, everything that AHA and ACC had written. And we said, where's the evidence? That's sort of a, where's the beef thing, that old commercial. What is the evidence? And we used the guideline framework, which I think is very good. When chairing the NIH, I was in a situation where they tried to push us to a new technique of making recommendations. But I like this because we say that Class 1, where the benefit greatly exceeds the risk, should be done. 2A and 2B-- it's reasonable. And Class 3 should not be done, either because it's dangerous or it doesn't work. But with each recommendation, you have to qualify it as to whether there's any evidence to support it, whether there are some studies, observational, or whether it's an expert opinion. And there are situations where you just don't have the randomized trial. And the one that gets mentioned frequently is the jumping out of a plane without a parachute. But there are many other situations where you cannot randomize patients-- hypertrophic cardiomyopathy, the defibrillator-- placing it, when to do it. So there are a lot of situations in imaging. But this allows you to make a recommendation and then to say what is the level of evidence. And clearly, there is not-- and I don't think will ever be-- enough evidence to go around to recommend what we should do on all the patients we see. What we came up with was that there are basically, at this point in time when we published in JAMA, only four guidelines that had more than one out of five recommendations that were strongly based in evidence-- heart failure, secondary prevention, acute coronary syndromes, and CABG if you give them a break. The majority, roughly one out of 10, had randomized trials that would support what was recommend. But I do want to call your attention to the fact that if you look at secondary prevention and prevention in general, which is where we are today, what we're going to talk about, that the dark blue is the Class 1 evidence. And this would be B. So A and B combined, more than 90% of recommendations for prevention, have some evidence base, either observational trials or randomized clinical trials. So this is an interesting area to get into. Shortly after this, the publication of the JAMA article, I got a call from Betsy Nabel, who was then the head of NHLBI. She subsequently has gone up to Brigham to be the CEO up there. And she asked me if I would do her a favor. And I really liked Betsy Nabel a lot. For you, Betsy, anything. What do you want? She said, I'd like you to share the new NIH guidelines. They're going to be evidence-based, like the Institute of Medicine wants. We're going to have six of them. We've got 90 people. I'd like you to chair it. And I said, Betsy, do you realize that I've just published in JAMA saying we don't have enough evidence? And you're asking me to chair this group on evidence-based-- are you sure you want me to do that? Yes, you're the one. We have $22 million. And we want to do it the way the Institute of Medicine says it should be done. So I agree to do it and went up there. And I'd like to call your attention to the fact that when you start looking at RCTs, there are situations-- and every time you look at a study, ask about this. First of all, the representation of women-- frequently with coronary disease, it's not that we have anything against women. It's just if you look at new drugs coming to market in coronary disease, many of the earlier studies exclude older patients-- co-morbidities, they've got other things. And so you'll get a population of younger patients, mean age of 62, and that's going to be roughly 75% men. Women don't have their heart attacks until they're a little bit older. So frequently, the representation of women is not there in terms of an evidence base. The elderly-- many of the trials exclude elderly. And of course, the bar for elderly keeps going higher as I get older. But still, you get up into the 75, plus age group, it's hard to find really good studies that look at that. And so we end up extrapolating something that works in a 58 year old into populations of people that are between the age of 80, 90, maybe even up to 95. And that's not necessarily correct. Racial and ethnic groups-- I'm doing a lot of work now internationally in China, Brazil, Chile looking at the development of guidelines just on a study on dual antiplatelet agents. The Chinese are more likely to bleed. The evidence base in inclusion of East Asian and South Asian is very small in our studies. So what do we know about dual antiplatelet agents after PCI in China, the largest country in the world-- 1.3 billion? Very little. Are they going to bleed on prasugrel and ticagrelor? We know how to use clopidogrel over there. So look and be sure that there is an inclusion of racial and ethnic groups. Even the new risk calculator that we have includes African-Americans. But we're trying to upgrade it so that it will predict in our Asian population. Multiple co-morbidities-- there's no sign on my door that says, Doctor Smith only sees patients that do not have renal failure, that have no chronic obstructive pulmonary disease, and are under the age of 70. They come in, I have to see them. Many of the trials, the evidence base has exclusion criteria that eliminate patients that we need to see. So whenever you see an RCT, start asking the basic question, who was included in this study? And be careful, because the press will give you a sound bite-- new study shows this. And it's assumed that it works in everybody all the time and that everybody's been included. So with that background, here's what we had. The last blood pressure recommendations were JNC 7 in 2003. The ATP III Update was 2004. Obesity, the largest problem in the United States-- no pun intended-- 1998. So all of these guidelines were a little rusty in terms of having been updated. The process that we used was to develop six panels. There's one that's coming out on implementation. What's the science behind implementation? How much do we know about what works and what doesn't work? Each panel was given the opportunity to develop critical questions. And so we had blood pressure, cholesterol risk assessment. And there's a new risk predictor out. And also lifestyle-- what should we eat, how much should we exercise, what do we really know about lifestyle? Shortly after we were told to develop critical questions, we were told that even though we had $22 million, there was not enough money to answer all these questions, to do systematic reviews. So everybody had to go back and cut it down to three critical questions, with the exception of obesity, which got five. Now, at this point in time, Michelle Obama had decided that obesity was going to be an important area. And I became aware gradually-- and that continued over five years in NIH-- that there are some discussions that go on between the NIH and the White House. At any rate, the obesity panel got five critical questions. The rest of these groups-- I've never seen so many semicolons and commas appear. These are really smart people on the panels. They did a good job of trying to broaden their critical questions. The important point is that these guideline statements--- and we're going to talk about blood pressure and cholesterol-- are limited in terms of the systematic review. They are not comprehensive. They never were meant to be. So there will be things that we would like to have included. Ambulatory monitoring of blood pressure-- what do we know about that? How should that be done? The whole issue of triglycerides and things with the liver-- could not include those things because of the limitation in terms of doing systematic review. So the process to develop these two statements-- strictly evidence-based-- included primary care. For the first time, they insisted each committee has primary care, plus cardiovascular and other representation in the committee. So it's different from the other groups in terms of the makeup of the committee, the fact that they insisted on the use of only evidence-based medicine, randomized trials. The trials themselves were evaluated and rated independently. So the committee did not sit around and say which was a good trial and which wasn't. They put out criteria for evaluation. But the decision about what was good and what was bad was based upon criteria specified by the committee, but done independently. And again, I mentioned the three critical questions. About 4 and 1/2 years into it, we heard from Francis Collins that the NIH didn't want to publish any guidelines. So you can imagine the 90 people that had been on these teleconferences, some of them weekly, going on were not happy. And why this occurred is-- it was never clearly specified. A lot was going on. Mammography guidelines had come out saying five years, rather than four years. And there was an uproar about that. Where is this coming from? It's not right. Women should be able to be screened earlier. The government's trying not to pay. Sequestration was there. NHLBI was trying to figure out funding, what they were going to be doing and what they should be doing. There were many issues on the table. And this publication came out saying that NIH had decided not to publish the guidelines. American Heart Association, American College of Cardiology, which does guidelines, quickly jumped in and said, we will take the evidence base and move it forward. NIH said, we see our role as evidence synthesis. We think the collaborative partner should be the professional societies for guidelines. Because after all, they do involve judgment. And we see knowledge generation for gaps and implementation. So this is then moving forward. And so the new guidelines that came out were published-- 2013-14-- with four of them-- obesity, cholesterol, lifestyle, and risk assessment-- as ACC/AHA guidelines, including others. The hypertension committee said, we don't want to have any more review. In fact, we think it's incorrect to the process. We've had our process. We've reviewed the evidence. We want to publish where we are. So they published as a report of the committee organized as JNC 8 in JAMA. And in fact, the NIH has set up a new type of grading of recommendations. All of these had to be translated into the ACC/AHA form, with the exception of the recommendations in JAMA, which maintain the original new NIH guideline classification, which is like the grade classification. You will never see this again, because NIH just decided not to do clinical guidelines. But that's why there's a slightly different type of wording in the two guidelines. So I'll move into the cholesterol. These are the ACC/AHA guidelines. Neil Stone chaired it. He's a lipidologist, a cardiologist, trained at the Brigham NIH with Levy and Frederickson, now at Northwestern. Jennifer Robertson is an internist, does a lot of excellent work in lipids. And Alice Lichtenstein is maybe one of the best in the world on nutrition. So the leadership group was diverse, with a good background. I had an opportunity to serve on this. We had others-- Bob Eckel in diabetes. But across the board, a combination of internists and cardiologists. The first question that comes up has to do with the transition from observational evidence to randomized trials. And what goes on is pretty much what happens in life-- we see something happen. It disturbs us. It happens more times, one or two times more. And we begin to wonder, what causes this? And so based on observation, we draw conclusions. We see populations that have high percentage of coronary heart disease. And we observe that their LDL is high. And we see other populations with lower LDL and observe that they have fewer heart attacks. So we develop an observational base. It does not necessarily follow like the night, the day that lowering LDL cholesterol with a drug or with diet necessarily results in a reduction in coronary events. So we need to develop trials in a randomized, controlled manner that show us that. And so as we hook the trials in-- this is the 4S, the initial blockbuster with placebo and treatment and lipid-- we begin to see a trend that, in fact, using statins in this case, lowering LDL cholesterol does reduce events. The question is, just exactly how far should we go? We also observe that the benefit is related to the relative risk of the group upon which we intervene. And it's true, in most trials, if you want to do a trial, you want to start with high-risk people, because you're more likely to see a benefit. It takes longer in a lower-risk group. Primary prevention is down here. It's a tougher thing to show, large population, lower risk. But lowering of LDL does result in a reduction in event rate. As we go to secondary prevention and then up to diabetes, it's an even greater result. The question still remained and was before the committee about the level to which we should lower cholesterol. This is a linear phenomenon. It works so well. It feels so good. Should we just keep doing it and just keep driving it down lower? Or maybe there's a threshold. Maybe it's a magic number that humans have, below which lowering cholesterol will have no benefit. Or possibly it's curvilinear. That is at higher levels of LDL, we see a greater reduction than at lower levels of LDL. So the issue of target was right in front of the committee. Now, targets had been a changing target, if you will. If you think about ATP III, ATP III said LDL cholesterol over 130-- these people need to be treated. Treat them. If they're between 100 and 130, you have your choice-- diet or treatment. And if they are less than 100, no therapy seems to be indicated. Now, that was based upon the existing knowledge from RCTs, which had never tested an LDL target. And to this date we still have no randomized trial that is actually titrated and tested. Well, with this background, HPS came out. And they showed that if you treated patients with simvastatin, in this case-- 40 to 80 years old, high risk-- that you would see about a 25% reduction in events. So that was correct. They showed that similar treatment-- and this is 40 milligrams of simvastatin-- for those with 100 to 130 had a benefit. But the surprise was that if you took patients whose LDLs were less than 100-- 90, 85-- start them on simvastatin, anyway. It's lower than what was the recommended LDL. You would see a 25% reduction in events. High-risk patients or those with previous disease benefited from statin therapy. And certainly, 100 was not a magic number. And then along came PROVE IT-TIMI 22, which was not a study designed to necessarily test the target. What was going on at that point in time was that the Brigham had done a lot of work with pravastatin. Atorvastatin was a new gorilla-- 80 milligrams, obviously much stronger than prava. And a lot of the data had been based on what was learned from pravastatin trials. And so they wanted to see, what happens if you use a stronger statin? And so they took some ACS patients. And they treated them-- 80 versus 40. Now, serendipitously, the LDL in the prava group made it to 95. They weren't titrated. They just ended up at 95, which is nicely tucked in under the then-target of 100. The other group came out at 62. And there was a benefit from the more intense statin. So I was on the committee at that point. This was with Scott Grundy chairing. And we sat around the table. And we said, what are we going to do? It looks like you have greater benefit. There is a 100 target that's been specified. And we said, well, we've been going 160 for primary, 130, 100. Let's give them 70. That'd be another 30 off. Doctors can remember that. And so 70 got established. There was never any titration to less than 70 upgrading. It was just, seems like we're going to get better if we go lower. That's where 70 came from. So one of the critical questions in front of the committee was, what's the evidence for goals? Because these goals then fostered a whole industry of people. As Harlan Krumholz has said, we're pulling drugs off the shelf to achieve a target which had never been tested with drugs that had no clinical outcomes that had been verified. The whole thing that was driving it was the surrogate endpoint of LDL. So we had a whole industry of adding ezetimibe, different medications, to try to get to target. And I was practicing that way myself. And so it was really important to understand. And hospital systems were being measured by whether or not you achieved 70. How many of your patients were under 70? So people are getting run up the flagpole because they couldn't get their patient under 70. So it was really important for the committee to try to understand, where's the evidence here? Now at about the same time, Rory Collins independently, unbeknownst to us, over at Oxford was asking sort of the same question. And these are his data, meta-analysis from over 170,000 participants, 26 randomized trial. What they found was looking at comparison of high statin versus lesser statin or statin versus control, was that at any level of LDL-- so you have to convert this. Just multiply by 40 to make it easy. It's really around 39. But this would be less than 80. And you can see the benefit from either statin versus control or more statin versus less. Across the board, regardless of the level of LDL, down to 80, initiating statin therapy or increasing it held benefits for patients with disease or at high risk. They found that this benefit extended across major coronary events, coronary revascularization, and even stroke, with the exception of hemorrhagic. Patients with atherosclerotic vascular disease or at high risk benefited from reduction of LDL using statins. And more intense seemed to be more effective than the less intense. And also, it didn't matter what type of manifestation-- previous disease, diabetes, type 1 or type 2, male, female, age-- and again, it's a little weak up here in the older group, you can see. We need more data there. BMI, smoker, non-smoker-- they benefit if they have the disease or were at risk. In addition, they looked at the issue of whether or not patients in primary prevention would benefit and came up with a conclusion independent of our group that the current guidelines out there and the criteria for initiating therapy did not do a good job of predicting patients that would benefit. We were missing patients that would benefit from lipid lowering therapy. They suggested that the guidelines needed to be reconsidered. They pointed out in certain groups that there was 17% reduction in total mortality treating patients that might not have been included in the current guidelines. So looking at randomized clinical trial, goals for primary prevention, we again found no RCTs that supported a specific target and a lot of evidence that we needed to rethink the criteria for initiating therapy and the type of therapy used. That led to the new risk predictor, Global Risk Assessment. And what happened with that was we moved outside of Framingham. Remember, Framingham has been used for years as a predictor. But this is not a population that represents the beautiful mosaic of the US population now. African-American, East Asian, South Asian, Latino-- not that prevalent up there in Framingham. So what we did was to expand the cohort, including [INAUDIBLE]. So now we have a very robust white and black cohort. We also included stroke. So the thing to recall is that this new cohort actually predicts both heart disease and stroke and includes African-American and the Caucasian, the white group. What we're working on now is something that will predict accurately in East Asian patients, at least, working in China right now with the people from NIH and Framingham. There is a calculator that you can use. Most of you know it. It's on your phone. You can sit down with the patient and actually calculate the risk just from knowing gender, age. You plug this in, and it will give you some estimate, 10-year risk. It also will give you a lifetime risk, which was one of the concerns about the previous risk estimator, that will help you with the behavioral recommendations. So this is up to the age of 59 and is intended to drive discussions on diet, exercise, and smoking. So here's an example of what can happen. These are two women, one African-American, another white, the same numbers. The African-American woman would be recommended for treatment, largely because of the higher risk of stroke in this population. The white woman would not. So the use and inclusion of stroke and inclusion of African-Americans helps us begin to decide on patients that previously would have been excluded from therapy. When you reach a point-- and this is not to dictate what to do and when to do it-- but it is to stimulate a patient discussion. There are many situations where it's sort of on the line. And that's where we felt that the use of CRP, of CAC score, ABI, family history, all of these things, come into play. I especially think the coronary calcium score can be very helpful. There were discussions after JUPITER came out that everybody ought to have their CRP done. I think Eric Topol said at a cocktail party, everybody should know their CRP. So there was a lot of enthusiasm for this. The committee did not feel that-- and we have all around the United States-- in shopping centers, you can see the advertisements. It's usually in the sports section. You can go and get everything scanned. They'll scan your coronaries, your heart, your abdomen. They'll give you an EKG. They'll give all this in a big package to you and tell you to go see your doctor at Barnes and ask the doctor what you should do. And that can all be done for $200 or less. So there's a lot of scanning going on. The question is, when should you do it? Should everybody have it? It's expensive. What's its role? The committee felt that the initial assessment of risk could be done using epidemiologic evidence and that beyond that, once you got to a level of risk, that's where these came into play. And I use CRP, coronary calcium. And in these groups, I find it very useful. They've made a major push on lifestyle. This came from the lifestyle group. Largely a Mediterranean-style diet. They again focused on the importance of the clinician-patient discussion. Computers don't practice medicine. You can't plug this into a computer. Doctors practice medicine using the tools that we can give them. And these guidelines are one tool that can be helpful, we think. So we ended up with recommendations that over the age of 75, moderate-intensity statin should be used because of the lower evidence, more side effects. But for patients under 75-- high-intensity statins, those with FH, high-intensity statin, those with a high risk in diabetes-- and this'll be most of the patients with diabetes-- should be on high-intensity. But there were rumbling discussions in the background after Hefner's New England Journal paper that not everybody with diabetes needed to be placed on high-intensity statin. The committee felt that those patients that had a lower risk with diabetes-- treated with moderate-intensity statin. So that was a change. And then, finally, the assessment using risk for the use of moderate to high-intensity statin for primary prevention. And another important find-- these are the statins. It's basically atorva-, rosuva- for high intensity, and then the moderate shown here. And remember that simvastatin at 80 milligrams is restricted to only those that have been on it for more than 12 months. There were a lot of interactions between simvastatin and other medications that are used frequently-- amiodarone, frequently-- and also patients that get put on antibiotics and the interactions with simvastatin. So simvastatin dropped a notch in terms of its potential role. Atorvastatin, generic, rosuvastatin, strong, prava in patients with transplants-- gradually working our way through to understanding the importance of achieving a dose. I want to say, though, that the committee-- and this is frequently missed-- that it's important to read the guidelines. Don't read the newspapers. Read the guidelines. Take a look at them. There is a Class Ia recommendation for monitoring a lipid panel. It should be done initially. It should be followed up when they're on therapy and continued. So it's important to know. It helps you with whether or not the patient's taking the medications, whether they're getting a result. We should be roughly 50% on high intensity, and maybe even more important now with the PCSK9 inhibition data coming out. So we did not remove the important recommendation about following lipids. What we tried to do was to balance the existing scientific evidence with a discussion between the patient and the physician to come up with a decision. Now I want to go to the hypertension quickly here. I think we've got some time on it. Again, this committee was chaired by Paul James, an internist from Iowa, and Suzanne Oparil, a hypertension expert from Alabama, past president of AHA, and a number of people on it which included internists as well as cardiologists. I also had an opportunity to work on this committee. It was published in JAMA. For those that say guidelines are too long, I'll point out that this is 13 pages and nine recommendations. Easy reading. You want to go through it quickly and see what the evidence showed. Now, this slide actually is compliments of Andy Kates, with whom I really enjoyed working over the past several years, and carries with it the Barnes logo. It's a part of my talk. I love it. It points out what we said earlier, that, again, with hypertension, we have the increasing observational data in population, but the question of the treatment effect. We know from Framingham that every increased blood pressure systolic over 118 is associated with increased risk of stroke and heart attack. The question is, what's the evidence that treating makes a difference? When should we start treating? How should we treat it? And how low should we go? Those are the key questions. And you all know Franklin Roosevelt was negotiating the future of the Western world in Yalta with a blood pressure that was up around 280. We used to think the older you get, the higher your blood pressure should go. You add the age to whatever it was, 160 or something, and that would be OK. And getting a little older, maybe you'd need a little more pressure to keep things going up there. Now we know that that's not true. And so the question is, when do you start doing it? And how much? What we do know from meta-analysis is that the treatment of both systolic and diastolic hypertension has significant reduction in fatal and nonfatal events. This is mortality and nonfatal and fatal events for systolic plus diastolic. This is isolated systolic. So we know the treatment works. The big question is, how should we treat? And how much? Again, similar to what was going on with Rory Collins in the UK, the Europeans were churning away and looking at evidence. And these are the data that they came up with independent of our work. This is their work in the elderly with hypertension. And these are the trials. So this initiation of therapy of systolic here of around 172 treating to 150. There was benefit. So there are a number of randomized trials out there in the elderly showing that treatment of hypertension worked. None of them treated to less than 140, with the exception of JATOS, which was in the Japanese population, that started treatment at a mean of 147 and got it down to 138 and showed no benefit. So these type of data then were before the committee coming out of our result that the available evidence in the elderly population for treatment really was that getting them under 150 was pretty good, but not a lot of evidence that going to the currently then-specified target of 140 worked. And in fact, if you take one trial here-- we say, well, what happens if you get your patient to 148 and they're on five meds, they're getting out of bed a little dizzy in the morning, you're thinking about sending them to the cath lab to do renal arteriograms on them, they have resistant hypertension-- should you add a sixth medication and drive them to less than 140? The only trial out there about that further titration happens to be JATOS, which showed no benefit from further reduction. So these are the type of data that were emerging. Now, similarly, there was a target in diabetes of 130. And really, no evidence there coming out that treating a 130 had a benefit. It looked like 140 would be good. And the same for CKD. In addition, there were three major guidelines. ALLHAT had driven JNC 7. That's where the thiazide-like preferential treatment came from. And the JNC guidelines said, use diuretics as your first choice. Europeans said, as far as they're concerned, you can take any of them, anything that lowers blood pressure works. You can choose what you like. The British were not too happy with that. They said, we don't like diuretics, and beta blockers are really off the shelf. So there's a lot of disagreement about how to initially manage hypertension. One thing that everybody agrees upon is the benefit of lifestyle. And I'll just call your attention to this JNC 7 recommendation, which was ratified with JNC 8-- weight reduction, a diet that's a Mediterranean-style diet, reduction of sodium to at least 2,400 milligrams, or 1 gram per day, exercise, and a reduction of alcohol in patients with known hypertension. All have benefit. And I've had at least one leading cardiologist in the country today, nationally very well known, who's told me he has hypertension. And he's noted that with these lifestyle medication, he was able to get off of two out of three medications. So they can have a profound effect. They are frequently neglected, because it's hard to get people to change the way they live. Given a choice between changing the way people live or taking a medicine, most people will choose to take a pill and keep on doing what they've always done. And yet behavior changes at the foundation made progress. So these are the three questions-- where should we start treating? How low should we treat? And do the drug classes differ? The committee quickly said, we ought to start treating at a point where we think people should be treated below. Why wait 20 milligrams? Let it keep going up. If we say it's under 140, it ought to be under 140. So they merged these two goals. There were several trials that looked at diastolic pressure treating to less than 90. They all showed benefit. There's only one trial that looked at the question of how far below 90. That was a HOT trial. It compared 90, 85, and 80, 50 to 80 years, with diastolic pressures 100 to 115-- showed no benefits. So getting to less than 90 was really what we found. And these are the data from HOT. Really no statistically significant data here. So a diastolic pressure, achieving a goal of less than 90 was what was recommended. Now, the major trials, systolic blood pressure, these are the ones that were included. It's another important point about systematic reviews-- your inclusion criteria will drive what you get to use. So always find out what the inclusion criteria were. Because that may tell you why a trial that you think is really good is not there. So ask that question, what were the inclusion criteria? What were these people doing? These are the five trials that made it. And notice that JATOS made it. There were three trials that treated to less than 150. And they all showed benefit. The two trials that got to less than 140 showed no benefit in the elderly. So that then confronted the committee with a major issue-- we've been treating older patients to less than 140, at least saying they should go. And it's not that easy to do in some older patients. Well, what should be recommended? The committee said a very strong recommendation-- over the age of 60 to treat to less than 150-- and a corollary recommendation that frequently is missed. It's very important. And that is if they're over 60 and tolerating therapy well, it's OK to treat them to less than 140. This was a major discussion. Because we'd been making progress and stroke. The Institute of Medicine and NIH said to only use evidence. The lack of evidence doesn't mean that something's wrong. But the evidence available supported treating to less than 150. A number of us argued for this corollary, which you'll note is expert opinion, that at the end there, that if you could get them to less than 140, it would be OK. I'm on the other receiving end of that, seeing patients come down to the cath lab, where they want me to do renal arteriograms when we do their coronary and find out what's going on, because they're old and we can't get the blood pressure from 148 to 139. So I feel that the strong, militant recommendation to everybody that's older be under 140 is tougher, but that we definitely should strive to drive the blood pressure lower. So this one is frequently not mentioned. But it is in there. Now, what about diabetes? I mentioned already the existing trials show benefit to treating less than 140. But the one that treated to less than 130, the ACCORD trial, showed no significant benefit in the total composite endpoint. But it did show a reduction of stroke in patients treated to lower levels. So it's a pre-specified secondary endpoint. Remember this. Because in about five days, the SPRINT trial will be reported. And we know already from the preliminary release that it was terminated early because of a benefit to treating less than 120. And diabetic patients are not included in SPRINT. I'm going to be at a big meeting on Friday to discuss this. SPRINT trial shows that treating patients less than 120 has benefit. But it did not include diabetes. What are we going to do when we have the ACCORD trial saying you don't need to go that low, but we do have this pre-specified secondary endpoint? At any rate, that led to a change in the recommendation for diabetes to 140. So the LIFE trial took a look comparing losartan to atenolol and showed, actually, an increase in stroke in patients on atenolol. Similarly, it showed an increase in new onset diabetes. For that reason, the beta blockers were taken off the list for use in patients with hypertension, as the Brits have done, with the exception of those who have compelling indications, which are ongoing angina, arrhythmias, heart failure. There are a number of patients where beta blockers should and could be used. But as a first line agent, they were taken off as a medication of first choice. This raises another issue that troubles me. It's assumed that there's a class effect. But how many of you use atenolol as your first beta blocker? I don't see any hands up. And mine's not up. It's metoprolol, carvedilol. We're using newer generation beta blockers. So if the dictum is, thou shalt use evidence only, and the evidence comes from a medication that we're not necessarily using that frequently, what is the safety of a class effect assumption? At any rate, beta blockers have been moved off based on this. And I'm just sewing some seeds for you to think about in terms of what type of class effect may or may not exist. So the new recommendations are-- use diuretics, calcium antagonists, or ACE inhibitors, with the exception of ACE inhibitors in African-Americans, beta blockers only if there are compelling indications. Now, the African-American data came out here from ALLHAT. Going back through in the systematic review, we found that there was an increase with ACE inhibition of strokes in-- you can see the increase in strokes-- in patients treated with lisinopril, compared to Chlorthalidone. So that's another new observation that is not necessarily mentioned. So this is the schema that's in the article. Over the age of 18-- it's pretty simple. Because the goals now-- everybody should be treated to less than 140. Chronic kidney disease, diabetes, and just the general population under 60. If they're over 60, the evidence supports treating to less than 150 over 90 with the proviso that if you get it to less than 140, that's fine if they can tolerate it well. The choice of medications you've heard in the African-American population-- and this was, again, a surprise to me-- lisinopril should be used only if they have chronic kidney disease as a starting agent. So it's pretty simple. So before, we had 130, 130, 140, back and forth. It's a fairly simple recommendation. Now, this is the trial that's going to come out at AHA. And I think it'll be published either in JAMA or New England Journal of Medicine. We're meeting twice there. I have to go down there early for a day, cutting into everything. And we're meeting again Monday night afterwards. It's a multi-center trial where the patients were treated to less than 120, compared to less than 140. The inclusion criteria are patients over the age of 50, blood pressure greater than 130, and they had cardiovascular disease, kidney disease, or high risk. So this is not everybody. But you will see in the headlines, I predict, new study recommends patients be treated to less than 120. But it's not all people. These are older people that are at higher risk showed some benefit. The real problem is that patients with diabetes are excluded. History of stroke is excluded. And there also is another associated trial that's looking at what happens to older people whose blood pressure is under 120 for longer periods of time? What are the cognitive issues? That's not going to be available to us. That's not going to be out for another eight or 10 months. So one of the dilemmas from a guidelines standpoint is, what do you do? You've got this evolving database where certain key people-- diabetes, older patients, cognitive issues-- not out there yet. How do you put all this into some meaningful recommendations for people that need to take care of patients? But the results are pretty impressive from driving the blood pressure lower. In my presidential address at AHA, I made the statement that we really needed-- I had written a talk on the contributions of invasive and interventional cardiology to care for patients. And at that time, I was in the cath lab and interventionalist. I tore the talk up because the 4S study came out. I had been at NIH doing lipoprotein research earlier. And I found my life coming in full cycle. At last, the theory of lowering cholesterol had been proven, reduction of total mortality. And I realized that we had to do more than revascularize. And so I made the statement at the end that rarely has cardiovascular medicine faced greater challenges. Rarely have patients expected more of us. But never-- never-- have we had such tools to make people well. It's time to make good on that promise. And I strongly came down on the fact that we needed to do a better job with comprehensive medical therapy. And that led to a lot of the quality improvement programs. We now have over 21 hospitals in the United States practicing, using evidence-based medicine. 80% of our population lives within reach of one of those hospitals. There's been a 30% drop in 30-day mortality from myocardial infarction, 16% drop in congestive heart failure, 4.6% from stroke. So the use of evidence and the organization of hospital systems has been associated with substantial progress in the United States at a time when the rate of this is going up in the 80% of the world is developing nations. It's working. But they say, what do you think? It's been 10 years after that about evidence base. And I made the comment, "promises to keep, promises to keep, but miles to go before we sleep," drawing from Robert Frost, because I really do think that, as we said in JAMA, we need a lot more evidence and have a lot more work to do if we're going to do the best we can for our patients. So I would encourage all of you in your careers to continue to think, to continue to look for opportunities to be involved in research, clinical trials. Do something that goes along with your practice in medicine. And you'll find those opportunities everywhere. I was involved in writing a paper with Salim Yusuf, and Scott Grundy, and Valentin Fuster on just exactly what principles should we have internationally? We said that governments and national societies need to collaborate. This has led to the project I'm doing in China with 150 hospitals, to Brazil, Chile. We've got to have the government and national societies working together-- NIH, ACC/AHA. Evidence-based guidelines must incorporate professional judgment, we said. There isn't enough evidence. And the assessment of risk needs to be involved. The higher the risk, the greater the benefit. I want to call your attention to the fact that if you look at the RCTs available, predominantly they are available in North America and Europe. China has only 2% of the evidence in the world. It's the largest country. So this is where I'm living now, trying to get out beyond and looking at the challenge of the NCDs and this epidemic of cardiovascular disease. We need more evidence. If you look at JUPITER, a phenomenal trial-- if you look at JUPITER and you live in China-- 1.3 billion people-- what evidence do you have about what you should be doing in China? Anybody find China on this list? Nada. They didn't test anybody in China. So there's no evidence about what the Chinese population should do with CRP. The dual antiplatelet trial came out. I'm on a writing group on, what should we do? How long should we continue dual antiplatelet therapy? This is a big trial. And as you can imagine, I went right for the inclusion criteria, because I'm writing this dual antiplatelet statement over in China, Japan, and Korea. This is stuff I need. And I said, well, how many Asians are in this study? Nada. Nobody from India. Nobody from China, Korea, Thailand. So I talked to one of the investigators. And I said, what's the story? Where are the Asians? They said, well, we've included Australia. I said, wait a minute. They need to go back and look at geography. That's another continent. So we do have a lot to learn internationally. And we concluded saying, "Whereas the causes of cardiovascular disease are common, the approaches will vary for cultural, social, medical, and economic reasons." And I think that continues to be a challenge for us. So with that, I'll conclude. And thank you very much for your attention.