December 10th, 2013
Co-Chairs of CVD Risk Guidelines Discuss a Difficult Case
Eric Lindley, MD, Donald M Lloyd-Jones, MD, ScM and David C. Goff, Jr., MD, PhD
University of Utah cardiology fellow Eric Lindley presents a challenging case to the co-chairs of the new 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk: David Goff, Dean of the Colorado School of Public Health, and Donald Lloyd-Jones, Chairman of the Department of Preventive Medicine at Northwestern University.
Lindley: My patient is an asymptomatic 38-year-old white male who comes to the office concerned about his future CV risk. He is a lifetime nonsmoker and takes no medications. Asymptomatic and very active, he frequently rides his bike to and from work, and spends his weekends hiking and camping. By trade, he is a horticulturist, eats out of his garden in Oregon, and eats red meat approximately once a week.
The patient has a significant family history of early atherosclerotic disease. His father, uncles, and grandfathers had either an MI or a CABG when they were in their mid-to-late 40’s. His physical exam is unremarkable, with a normal BMI and blood pressure (systolic blood pressure <125 mm Hg). His total cholesterol has ranged over the last five years from 200-240, with his LDL <190, and his HDL hovering around 35-40.
He is reluctant to start a statin at this age, as he doesn’t want to start a medicine he’ll need for life unless the benefits truly outweigh the risk. Using the new lipid guidelines and risk calculator, his lifetime CVD risk is estimated at 50%. However, until he is age 40, no 10-year risk is calculated. Estimating for him in the future, and only changing his age in the calculator, leaving all other numbers the same, his 10-year CVD risk isn’t above 7.5% until he’s 53. Yet a significant number of his relatives have already had their events by the time they were 53.
Would you start a statin based on his lifetime risk alone?
Goff: I would be reluctant to recommend a statin based on his lifetime risk estimate alone, primarily because he expresses reluctance to take a life-long medicine at this point, he appears committed to a healthy lifestyle, and his short-term risk and short-term benefit are both estimated to be low. I would reinforce a healthy lifestyle and try to help him reduce his cholesterol further. You stated that his blood pressure is normal. If it were higher than optimal, that is, higher than 120 systolic or 80 diastolic, I would include an emphasis on the blood-pressure lowering benefits of more intensive lifestyle change to motivate adherence.
Lloyd-Jones: This is a very interesting scenario, and this is the kind of patient I encounter very frequently in my preventive cardiology clinic. First and foremost, we must respect his family history. This kind of premature CVD family history is not particularly well represented in any risk assessment approach from the broad population. He must be congratulated on trumping as much of his inherited risk as possible by following the healthy lifestyle he follows. He sounds like he is quite active and currently asymptomatic at a good workload. That he has stayed lean (“normal BMI”) is a good sign. If he wants to avoid medication, he should certainly be offered a referral to a nutritionist or dietician to explore options, explicitly including a vegan diet. But, at this point, I would not start a statin based solely on the lifetime risk estimate.
Would you wait until his 10-year risk exceeded 7.5% to start a statin?
Goff: Given his strong family history, I am concerned that the risk calculator might under-estimate his risk. All risk estimates are averages for people with the same values of the risk factors that are included in the model, but no one really has 7.5% of a heart attack or stroke. The patient will either have an event or not during the 10-year period. From that point of view, it is all or nothing!
The challenge is that it is difficult to predict who will have the event and who will not in a group with any specified level of risk, and when we examined the contribution to risk prediction of information regarding family history, that added value was quite small for the general population. Of course, most people don’t have the very strong family history of this patient. For him, I would be interested in discussing disease screening, especially if we were not able to achieve substantial reductions in his cholesterol through lifestyle change.
Lloyd-Jones: Let’s think about his risk. I would start by plugging in his risk factor levels with an age of 40. Recognizing that we have not quantitatively considered his family history, I get 10-year risk estimates in the range of 1.4% to 2.4% (if he were black it would be as high as ~3.1%). Lifetime risks are in the range of 46-50%. This is not where we prescribe a statin per the guidelines (despite what the media seems to think), this is where we start the risk discussion. And this is very much how I practice.
As you know, the guidelines provide 6 different things that should potentially be considered to color in our risk assessment. In this framework, the first 3 questions to ask, before we consider any additional testing, are:
- Is his LDL >160? Sounds like it sometimes is, so I would take this as an indicator that we should consider a statin now.
- Does he have a strong family history of premature CVD? Unequivocally, yes. This, for me as his physician, would provide strong justification for a statin now.
- Does he have a higher lifetime risk? I would say unequivocally, yes.
At this point, I would provide him with all of this information, answer any questions, and indicate that, on balance, I would favor starting a statin now. But first I would tell him my read on the statin safety data.
Would you do further testing such as an hsCRP or CAC score to determine his risk and possible need for a statin?
Goff: The Risk Assessment and Cholesterol Guidelines provide advice on how to proceed when one is uncertain about a risk-based treatment decision. We recommend that clinicians consider using information about several factors: family history of premature CVD, LDL-C > 160, lifetime risk, hsCRP, ABI, and CAC.
I am especially intrigued by the potential value of disease screening using CT to detect CAC. We know that people without CAC are at very low risk for clinical events, and people with CAC are at much greater risk. If he were interested in pursuing additional testing, I would consider getting a CAC score. Even without the CAC score, he has a very strong family history and a fairly high lifetime risk. If his LDL-C were to remain greater than 160 despite focus on a healthy lifestyle, the guidelines provide support for a decision to start statins. The risk-benefit discussion will be critical for this patient, because his short-term benefit is likely to be low, but his long-term benefit is likely to be great.
Lloyd-Jones: If he is still reluctant to take a statin, or would like to make our decision-making a little more concrete, here is what I would do next:
- Although these tests are mentioned for potential consideration in the guidelines, I would not personally measure CRP (too nonspecific) or ABI (very low yield at this age) in him.
- Off guidelines, I would next offer him a Lp(a) measurement. If that is high (>30 or so; which I suspect will be the case, given his family history), I would urge statin therapy. Lp(a) is actually our highest yield test for heritable premature CVD and I believe it would indicate intensive statin therapy if high in this scenario.
- If the Lp(a) is not elevated, or he is not convinced enough to start a statin in the instance that Lp(a) is high, I would offer him a CT scan for CAC scoring. This would be an important test, in many ways definitive for decision-making in such a scenario. In fact, this is the situation when I tend to use CAC scores. The average white male develops CAC at an age of 53 years, so if he has CAC now he is way ahead of his age/sex/race cohort and it would indicate a sufficient burden of coronary atherosclerosis to consider him at very high risk in the nearer term. If his CAC score is zero, this would be very reassuring that he should be at very low risk (<1%) for the next 5-10 years. I would feel compelled to tell him that he will have to pay out of pocket for the CAC score (typically $75-$300) and that there is some radiation exposure, typically equivalent to ~2 chest x-rays). If we can buy him 5 years, there might well be other medication options for risk reduction other than statins.
If, after all of this, he still does not want to take a statin, I would at least feel I had offered him the best possible information to help him make his decision. Regardless of his decision, optimal lifestyle modification must remain a critical component of his prevention regimen.
8 Responses to “Co-Chairs of CVD Risk Guidelines Discuss a Difficult Case”
NEJM Journal Watch — Recent Cardiology Articles- Anticoagulation in Patients with Cancer-Associated Pulmonary Embolism: How Long is Long Enough?
- Spironolactone Use After Percutaneous Intervention for Acute Myocardial Infarction
- CRISPR/Cas9 Gene Editing for Transthyretin Cardiomyopathy
- Does Colchicine Reduce Cardiac Events After Myocardial Infarction?
- First-Line Therapy for Ventricular Tachycardia After MI: What's the Best Approach?
I agree with all the comments. In this case the benefits of Statins use truly outweigh the risk? No, even better answer, we don’t know. He is doing everything right, asymptomatic, excellent quality of life. Why we want to fix something that is no broken? Why every time we have a knew guide or calculator we use them like the bible, like is the source of all the knowledge, like they are perfect. We use statistics and risk calculation like the Pythia use to do in her consultations in Delphi. In this case the patients is more intelligent, he is doing the right think and asking the correct questions. I think he already know the answer.
When the panel of the AHA/ACC developed the risk calculator they review the existent calculators, all of them were imperfect, so they developed a new one. In this new calculator the family history is not included, why? Is it not a good risk factor predictor? When they tried to perform an external validation they were not good. Do we have studies comparing this risk calculator to the other or the ability not only to predict risk accurately but also affect patient outcomes? Why the treatment threshold is 7.5%?
In this particular case the imperfect risk calculator doesn’t apply, this particular patient is in a very special subgroup. Where is the treatment threshold, we don’t know. If you decide to treat, where is the lipid level target, we don’t know. Are we sure the use of statins for very long time is going to harm the patient? We don’t know. Do we know what is best for the patient in this particular case, except for the things he is already doing. No.
What happened with the pledge “do not harm”. If we are going to use a statins in massive proportions because we think theoretically is sound and we will be careful monitoring side effects, and we will wait for future validation of our risk assessment and how this affect patients outcomes is simply wrong.
If you tell this particular patient your case is very special, your future is not written. At this moment with the information we have we don’t know when to start statin therapy, how much to give you, we don’t know if it is completely safe to use them for very long time and if we are going to change anything, but theoretically could help you and we are going to be carefully monitoring for side effects you can try it. If you want I can try to obtain more information with the test already described above, and then let the patient decide which is the correct pathway.
I also completely agree with the above. We also do not know whether the relative risk reduction with statins increases with longer duration of use. This could change the risk-benefit relationship, as lifetime risk could be reduced not only by covering more years with a statin but by incremental relative risk reduction the advancing age. “Mendelian randomization” in PCSK9 variants support this as mutations yielding a relatively small percent LDL reduction (say 20%) seem to yield a robust long-term reduction in CV events ( say 50%), similar to the reduction we would expect from robust LDL reduction with potent statins.
I would like to make few points:
1-There is a very strong family history.
2-Young person is facing this risk.
3-We cannot claim our risk prediction is infallible and so
4-grab at anything likely to reduce the risk and therefore
5-start statins and watch carefully for side effects.
6-Tell the patient he has many more years to live and to
follow further developments in this field carefully which
will unravel methods to help persons like him.
Gads –Given the patient’s striking family history and the low lifetime risk of statin therapy, I think – and will accede to the comment that my recommendation may justify my assessment of a”no brainer” strategy. With monitoring of issues related to statin therapy, is there really a down-side to using a statin? And would it not be more feasible to measure lipid parameters than vascular progression, given that lipid parameters have been shown to suggest, if not guarantee vascular progression? Whereas repetitive vascular studies, if not bodywide, are expensive, reveal only macrolesions or “suggest that progression is occurring”, and are “late to the party., With monitoring of the labortory issues related to statin use, now largely minor, what would – except for expense — be the downside? And if the patient lived to be 100 + years old, with intact limbs instead of a fem-pop bypass (as a simplest case scenario), what would be the harm? Cost — oh, give me a break! Getting folks to decrease their dietary fat intake would accomplish a decrease in food costs, but would be much harder to achieve.And lipid testing is not as costly as daily elevated pathologic lipd generation from ingested sources/. Yes — dietetics matterss, but for a population of lean people, continued risk reduction via medication my be useful and life/limb-prolonging.
Gads –Given the patient’s striking family history and the low lifetime risk of statin therapy, I think – and will accede to the comment that my recommendation may justify my assessment of a”no brainer” strategy. With monitoring of issues related to statin therapy, is there really a down-side to using a statin? And would it not be more feasible to measure lipid parameters than vascular progression, given that lipid parameters have been shown to suggest, if not guarantee vascular progression? Whereas repetitive vascular studies, if not bodywide, are expensive, reveal only macrolesions or “suggest that progression is occurring”, and are “late to the party., With monitoring of the laboratory issues related to statin use, now largely minor, what would – except for expense — be the downside? And if the patient lived to be 100 + years old, with intact limbs instead of a fem-pop bypass (as a simplest case scenario), what would be the harm? Cost ? — oh, give me a break! Getting folks to decrease their dietary fat intake would accomplish a decrease in food costs and lipid assaults, but would be much harder to achieve. And lipid testing is not as costly as daily elevated pathologic lipd generation from ingested sources/. Yes — dietetics matters, but for a population of lean people, continued macrovascular risk reduction via medication may be useful and life/limb-prolonging.
I will try to keep this very short and sweet. If I were the patient I would try to do everything I could to prevent atherosclerosis. The only risk factors he has are an elevated cholesterol and a very strong family history. A randomized controlled trial is not focused on this guy. He is an n=1 study. I would prescribe a potent 3-HMG CoA reductase inhibitor to him. I think this would lower his personal risk of early MI substantially.
I’m surprised no one has really mentioned giving the patient a good idea of the benefit and harm and then letting the patient decide.
Give him a rough ballpark of his 10 year CVD risk – let’s say 8% or whatever it is – if he has a positive family HX then maybe increase that risk slightly
Statins reduce this by roughly 25% – so his risk will go down to 6%
1 in 10 will get muscle aches – 1/10,000 severe muscle kidney damage
The cost etc – based on the most recent guidelines tell him not to obsess with his cholesterol numbers.
And then support him in whatever decision he makes. It can always be revisited in a number of years.
If anyone is interested I have developed a one page website that I hope is helpful/useful that provides all that information so you can figure out benefits and harms in under 60 seconds – cvdcalculator.org. Would love feedback.