Author + information
- S. Matthijs Boekholdt, MD, PhD,
- G. Kees Hovingh, MD, PhD,
- David D. Waters, MD, PhD,
- Scott M. Grundy, MD, PhD and
- John J.P. Kastelein, MD, PhD∗ ()
- ↵∗Department of Vascular Medicine, Academic Medical Center, P.O. Box 22660, Meibergdreef 9, Amsterdam 1100 DD, the Netherlands
We read with interest the letter by Drs. Mascitelli and Goldstein commenting on our recent paper (1), in which we showed that the linear relationship between levels of low-density lipoprotein cholesterol and cardiovascular events extends to levels <50 mg/dl. Our findings are strongly consistent with the concept “the lower the better,” but translation of these findings into clinical practice is still challenging. Important reasons for this are the lack of trial evidence for a lipid target-guided strategy and the lack of evidence-based effective lipid-lowering therapies beyond statins. The issues raised by Drs. Mascitelli and Goldstein are less relevant.
Drs. Mascitelli and Goldstein noted that the primary prevention J-LIT (Japan Lipid Intervention Trial) showed an increase in risk of mortality among patients achieving a total cholesterol (TC) level <160 mg/dl. An important limitation of J-LIT is that it was not a placebo-controlled trial. If J-LIT had a control group not on statin therapy, it may well have shown a similar trend. Of the 28 patients in J-LIT who achieved a TC level <160 mg/dl and died, 12 died of malignancy. This association is caused by confounding, that is, (subclinical) malignancies decreasing cholesterol levels (2,3). Large-scale evidence from statin trials shows that lowering of cholesterol levels does not increase the risk of malignancies (4). In addition, in J-LIT, the statistically significant increase in risk of cardiac events among those achieving a TC level <160 mg/dl was on the basis of just 4 events, and the nonsignificant increased risk of stroke was also on the basis of just 4 events. In our meta-analysis, 351 fatal cardiovascular events occurred among those achieving a TC level <160 mg/dl, which makes our results statistically more reliable.
Drs. Mascitelli and Goldstein also pointed out that revascularization is a subjective outcome measure. Although the decision to perform elective revascularization may indeed be subjective, associations with risk factors tend to be in line with “hard” cardiovascular outcomes. Still, for 7 of the 8 trials, elective revascularizations were not included in the pooled outcomes of our meta-analysis; in the TNT (Treatment to New Targets) trial, elective revascularization for stable angina and unplanned revascularization for unstable angina could not be separated, and therefore both were included in the pooled outcomes (5).
Finally, Drs. Mascitelli and Goldstein stressed the fact that myopathy is not a trivial adverse effect of statins in clinical practice. We agree with this remark, as discussed extensively in our paper. Unfortunately, data on adverse effects and all-cause mortality were not obtained from the individual trials for the purpose of this meta-analysis.
In conclusion, the results of our meta-analysis are largely consistent with the concept “the lower the better” and warrant new trials to formally test this hypothesis.
- American College of Cardiology Foundation
- Boekholdt S.M.,
- Hovingh G.K.,
- Mora S.,
- et al.
- Jacobs D.,
- Blackburn H.,
- Higgins M.,
- et al.