Published in the August 2004 issue of Today’s Hospitalist
While a new study on the use of statins in inpatients may give hospitalists relatively few new answers about whether they should start inpatients on the drugs, it provides some interesting results about the importance of continuing to give statins to patients already taking the drugs upon admission into the hospital.
Investigators examined nearly 20,000 patients with acute coronary syndrome enrolled in the GRACE study from April 1999 to September 2002. Just over 4,000 of those patients “21 percent “were already taking statins when they entered the hospital. Eleven percent of those patients did not continue to receive statins during their hospitalization.
While previous studies have looked at the effect of statins on inpatients, this recent study, which appeared in the June 1, 2004, Annals of Internal Medicine, helps round out the picture for hospitalists. Investigators looked at three groups of patients: individuals who never took statins, patients who were hospitalized while already using statins, and statin use by patients who started taking the drugs in the hospital.
Researchers concluded that statin therapy before hospitalization affected the severity of those patients’ presentation and improved their outcomes. While patients already using statins presented with more comorbid conditions, they were less likely to experience a large infarct, have important clinical complications, or die during their hospitalization.
Researchers found that patients previously taking statins were less likely to present with ST-segment elevation or develop peak creatine phosphokinase levels more than twice the upper limit of normal. These patients were also less likely to die during hospitalization or develop selected clinical complications.
Of the 15,000-plus patients not taking statins when they entered the hospital, 38 percent began receiving the drugs. Investigators found that statin-naÃ¯ve patients who began taking the drugs as inpatients were less likely to die in the hospital than patients who never received the drugs. Investigators also concluded that starting statin-naÃ¯ve patients on the drugs when they enter the hospital for acute coronary syndrome may be associated with decreased hospital mortality.
Perhaps most interestingly, investigators examined what happened when patients previously taking statins stopped using the drugs during hospitalization for acute coronary syndrome. Much of the benefit of the statins appeared to vanish during hospitalization, with those patients experiencing an event rate almost three times that of patients who continued with their therapy. The article talks about a rebound effect that can hurt patients who stop taking statins, particularly when they’re experiencing an acute coronary illness.
The study adds to a growing body of knowledge about inpatient use of statins. During recent years, a number of other studies have similarly demonstrated the potential of statins to lower mortality and length of stay in patients with acute coronary syndrome.
This latest research, however, is unique for two reasons: First, it examined a large number of patients. Second, it looked at patients who were and were not already using statins when they were admitted to the hospital.
As the article and an accompanying editorial point out, the study’s observational nature raises questions about many of its conclusions. While lead investigator Frederick A. Spencer, MD, director of the coronary care unit at the University of Massachusetts Medical Center, acknowledges that the results by themselves may not warrant a change in practice, he says that they add to the growing body of knowledge about the value of statins in inpatients.
Today’s Hospitalist talked to Dr. Spencer about the study’s findings and the case for starting “or at least continuing ” statins in inpatients with acute coronary syndrome. Even when patients “fail” statin therapy, the drugs
He also talks about the challenges that researchers face in demonstrating the value of statins in the inpatient setting.
How does this study change or add to our understanding about statins in inpatients with coronary problems?
It’s a very small addition to the knowledge base already out there. Like the PRISM study, our data showed that patients coming to the hospital on statins seemed to have some protection with respect to hospital outcomes. We went one step further, however, and showed that if statins aren’t continued, patients lose some of that effect.
The more interesting result was that people coming in on statins had a very different kind of event. They tended to have smaller MIs or just unstable angina, and they had less hospital mortality, which is interesting. It suggests that even if you “fail” your statin therapy and show up with an MI, there is still a benefit to the drugs.
A similar question was raised with aspirin about a decade ago that’s still going back and forth, and that’s the issue of aspirin resistance. Are people who failed their aspirin therapy sicker because they are aspirin resistant, or is the aspirin still giving them some benefit, and as a result they have smaller MIs?
This study enhances previous trials, like the PRISM study, because it’s larger. It’s also a community-based study. While patients enrolled in clinical trials tend to be cherrypicked and are perhaps younger and healthier, patients in the community setting include all comers. So in some ways we would say that as flawed as these data may be, they may also be more generalizable to what a physician treating patients would see.
How does your study’s methodology limit the results?
As we pointed out in the Annals article, this study was meant to be hypothesis-generating, so it raises questions. The data are in keeping with some of the suggestive data that are already out there, but it really can’t be more definitive than that because it’s an observational study.
Imagine that a group of patients enter the hospital on statins, and half continue with their therapy and half don’t. Physicians will tend to stop or forget about statins in those patients who are really sick, like people who are on a ventilator, so the data could be pointing to an effect instead of a cause. We tried to control for that effect with multivariate analyses, but the data are still limited.
One of the reasons I felt OK about this as an observational study, even with its flaws, is that there’s not likely to be a randomized trial where patients come into the hospital with a heart attack and half are continued on their statins while half stop them. No institutional review board is going to approve that study. You can say that you’re a little nervous about continuing statins in the setting of an MI, but these data suggest that it’s important to continue the drugs.
I should also point out that the data on long-term adherence suggest that it’s important to continue these drugs. We know that patients who are on statins in the hospital with an MI are much more likely to be on them in a year.
You talk about patients who stop using statins in the inpatient setting experiencing a “rebound effect.” How might this affect hospitalized patients who actually stop taking statins?
We know that statins aren’t just cholesterol lowering drugs, that they do a lot of other things like reduce inflammation, effect endothelial function and relax the lining of the arteries. One body of work I cite in the article found that if you put a patient on statins, endothelial function improves markedly and your arteries are less likely to constrict. But when you take the drug away, there is a rebound, and you’re worse than you were at baseline.
That’s interesting when you think about someone coming in with an MI. They’re in a hot period when the plaque has ruptured and the artery is irritable. The biggest concern is death from the initial MI or a recurrent event. So you can imagine that if you have someone taking a statin and you take that drug away and they experience more spasm or arterial constriction a couple of days down the road, that could lead to recurrent events.
We didn’t look at the timing of that rebound, but the PRISM researchers did, and they found that it was in the first seven days after discontinuation.
How strong is the case to start statin therapy early?
That’s what the MIRACL study hoped to show, and the researchers were a little surprised that the results weren’t more impressive. It was one study, and maybe it wasn’t big enough.
It’s hard to show some of these effects because physicians have become really good at treating these patients. We’ve got them on aspirin, we’ve got them on clopidogrel, we’ve got them on beta-blockers and ACE inhibitors, so it’s getting harder and harder to show the incremental benefit of each agent. It’s harder to sort out which is contributing.
The data from the MIRACL study aren’t perfect, but they are suggestive. Then you have data from studies like mine, which suggests that it’s very important to continue patients on statins.
There are also data from registries that clearly show patients who start taking statins in the hospital are much more likely to be taking those drugs in a year, and that they’ll have a much lower mortality in a year. I think it’s important to consider this long-term adherence perspective.
An accompanying editorial in Annals says the major question is whether to start statins in the hospital or wait until later. How did your study help answer that question?
Because most of the data are suggestive, we don’t know. I would point out, however, that because of the accumulation of all these data and the various data on long-term adherence, groups like the American College of Cardiology and the American Heart Association appear to be moving toward a position where they encourage the early use of statins. We know that if patients get out of the hospital without being on a statin, they’re much more likely to never get it.
What kind of research would you like to see to help clarify the role of the inpatient use of statins?
You could take a large number of patients and start one group of patients on statins in the hospital, and give the other group of patients statins two weeks later. I think everyone would be comfortable with that, but it would have to be a very large study because all of the other medicines we give these patients are doing such a good job of lowering mortality.