How Low Should You Go? Is Very Low LDL-C Safe?

As clinicians, we know that lowering low-density lipoprotein cholesterol (LDL-C) with statins alone or in combination with nonstatin therapies such as a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor (eg, alirocumab, evolocumab) provides cardiovascular (CV) benefit. Patients on combination lipid-lowering therapy commonly present with LDL-C levels < 25 mg/dL (< 0.6465 mmol/L), and in clinical trials, extremely low LDL-C (< 10 mg/dL [< 0.2586 mmol/L]) has been observed. These findings have led many clinicians to question whether very low LDL-C affects formation of cell membranes, vitamin synthesis, and production of adrenal hormones.

To address the safety of very low LDL-C (defined in this article as < 25 mg/dL [< 0.6465 mmol/L]), we will review the challenges of obtaining an accurate LDL-C estimate; cholesterol metabolism; effects of low LDL-C concentrations in cord blood; genetic mutations associated with low LDL-C; study data on the effects of statins and PCSK9 inhibitors on adrenal steroids, gonadal steroids, and vitamin levels; and adverse event reporting in participants with low LDL-C in randomized controlled trials.

Obtaining an Accurate LDL-C Estimate

The challenges of accurately estimating LDL-C add to the complexity of assessing the safety of very low LDL-C levels. In some patients, the commonly used Friedewald equation underestimates the actual LDL-C level. Although it has been well established that the Friedewald equation has unacceptable accuracy in patients with triglyceride (TG) levels ≥ 400 mg/dL (≥ 10.344 mmol/L), a study by Martin and colleagues found significant inaccuracies at much lower levels (eg, TG levels of 150-399 mg/dL [3.879-10.318 mmol/L]). The study found that Friedewald equation–estimated LDL-C levels < 70 mg/dL (< 1.8102 mmol/L) may be inaccurate in patients with TG levels ≥ 150 mg/dL (≥ 3.879 mmol/L). In about 40% of patients with TG levels of 150-199 mg/dL (3.879-10.318 mmol/L) and 59% of those with TG levels of 200-399 mg/dL (5.172-10.318 mmol/L), the actual LDL-C level may be ≥ 70 mg/dL (≥ 1.8102 mmol/L). A major concern with underestimating LDL-C is that it could lead to withholding CV risk-reducing statin and nonstatin therapies in vulnerable patients.

One way to estimate LDL-C more accurately is to shift from a one-size-fits-all equation to a tailored one. Though more than 20 other equations have been proposed, Seth Shay Martin, MD, MHS (co-author of this article), developed a new equation, the Martin-Hopkins equation, which provides the best accuracy according to head-to-head comparisons and is recommended by guidelines around the globe. [Editor's note: Please refer to the following guidelines: 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines; Quantifying Atherogenic Lipoproteins for Lipid-lowering Strategies: Consensus-based Recommendations from EAS and EFLM; World Heart Federation Cholesterol Roadmap 2022; Lipid Measurements in the Management of Cardiovascular Diseases: Practical Recommendations a Scientific Statement from the National Lipid Association Writing Group.]

The Martin-Hopkins equation can be easily implemented in laboratory systems; this has already been done at individual institutions and companies, such as Johns Hopkins and Quest Diagnostics, and by entire countries, such as Brazil. There are no intellectual property restrictions or fees associated with use of the Martin-Hopkins equation.

Regulation of Cholesterol Metabolism

One concern with very low LDL-C levels is that they could result in a lack of available cholesterol needed for synthesizing cell membranes, steroid hormones, and bile acids (necessary for absorption of fat-soluble vitamins from the intestine). These concerns, however, are misplaced because evidence has shown that low cholesterol levels trigger cellular pathways that lead to increased cholesterol synthesis.

The liver synthesizes and secretes very low-density lipoprotein (VLDL), which carries TGs and fat-soluble vitamins to various tissues. The LDL particle is formed in the circulation by hydrolysis of TGs in VLDL, by lipoprotein lipase. The LDL particle may be removed from the blood by the hepatic LDL receptor or may enter the vascular endothelium, where it is the precursor of the atherosclerotic plaque.

LDL that is bound to the LDL receptor enters the hepatocyte and is transported to lysosomes, where cholesteryl ester (within the LDL particle) is hydrolyzed, releasing cholesterol for physiologic needs and recycling the LDL receptor. Statins act in the liver to inhibit 3-hydroxy 3-methylglutaryl-coenzyme A (HMG-CoA) reductase in the cholesterol synthesis pathway, thus upregulating LDL receptors and thereby reducing plasma levels of LDL-C. The LDL receptor on the surface of the hepatocyte may also bind to the protein PCSK9, which directs the LDL receptor to the lysosome, where it is degraded, leading to reduction in LDL receptors and increased plasma LDL-C. Medications that inhibit the action of PCSK9 (monoclonal antibodies to PCSK9) or reduce its synthesis through genetic mechanisms (inclisiran) reduce plasma LDL-C.

LDL-C Levels in Cord Blood

Other evidence for the safety of very low LDL-C comes from studies of cord blood. These studies have found low LDL-C levels in cord blood that increase within a few days after birth, showing that very low LDL-C in the placenta and umbilical cord are compatible with fetal growth in late pregnancy. In 36 control cord blood samples, mean LDL-C (± SD) was 36 ± 6 mg/dL (range, 18-46 mg/dL [0.4655-1.1896 mmol/L]). Mean LDL-C was higher in cord blood samples from 12 newborns at risk for type II hypercholesterolemia: 62 ± 16 mg/dL (range, 43-92 mg/dL [1.112-2.379 mmol/L]).

Another study of cord blood reported mean LDL-C levels of 22 ± 10 mg/dL (with an increase to 50 ± 19 mg/dL in the infant 4 days after birth).

Low LDL-C Due to Genetic Mutations

Genetic evidence has further assured the safety of very low LDL-C levels. Genetic mutations that cause abetalipoproteinemia, hypobetalipoproteinemia, and familial combined hypolipidemia are associated with reduced synthesis of chylomicrons, VLDL, and TG. Although these inherited diseases are associated with low LDL-C, the adverse effects are probably not due to low LDL-C per se, because secretion of other apolipoprotein B–containing lipoproteins is markedly reduced.

Abetalipoproteinemia is a rare autosomal recessive disease (1 in 1,000,000 persons) that is caused by mutations in microsomal TG transfer protein (MTP). MTP facilitates the transfer of TG into the nascent apolipoprotein B particle in cells in the intestine and liver, leading to the formation of chylomicrons and VLDL. Mutations in MTP reduce the transfer of TG and, as a result, reduce the production of chylomicrons and VLDL, leading to extremely low plasma levels of TG and cholesterol (< 30 mg/dL [< 0.7758 mmol/L]) and undetectable levels of LDL-C and apolipoprotein B in plasma. Patients with abetalipoproteinemia may have hepatic steatosis (fatty liver).

Familial hypobetalipoproteinemia is caused by mutations in the gene for apolipoprotein B, leading to decreased synthesis of VLDL and reduced LDL-C concentrations in plasma (< 30 mg/dL [< 0.7758 mmol/L]). Patients with homozygous and heterozygous mutant alleles have increased risk for hepatic steatosis, chronic diarrhea, steatorrhea, and deficiencies of fat-soluble vitamins.

Familial combined hypolipidemia is characterized by loss-of-function mutations in the gene for angiopoietin-like 3, which results in reductions in LDL-C, HDL-C and TG. A pooled analysis by Minicocci and colleagues found that although some patients with familial combined hypolipidemia may develop fatty liver, the prevalence was the same as that of the control group. Moreover, this analysis found that other patients were reported to be healthy. As mentioned earlier, the presence of multiple lipid abnormalities suggests that low LDL-C alone is not responsible for adverse effects in these genetic disorders.

Mutations in PCSK9

Individuals born with nonsense mutations in PCSK9 have lifelong low LDL-C levels conferring CV benefit. Patients with mutations interfering with PCSK9 have LDL-C levels well below the usual population average. They also have a reduction in CV events beyond what would be expected from clinical trials using LDL-lowering medications. This is probably explained by the long-term cumulative reduction in LDL-C starting at birth, driving lower events compared with the shorter time windows of clinical trials.

In addition to the benefit of low LDL-C, patients with loss-of-function mutations in PCSK9 have been found to be generally free of any significant adverse effects due to lifetime lower LDL-C levels. A UK Biobank study found that loss of function mutations in PCSK9 were not associated with increased A1c or glucose, hepatobiliary adverse events, or neurocognitive dysfunction. Analysis of two loss-of-function mutations in PCSK9 in the Quebec founder population showed no effect on the prevalence or age of onset of Alzheimer's disease or on neurocognitive function.

Studies Evaluating Steroid Hormone Synthesis

Despite concern that low levels of LDL-C would diminish steroid hormone synthesis, data from randomized clinical trials have found that reduction in LDL-C by lovastatin, pravastatin, and simvastatin and reduction of LDL-C to low levels by evolocumab and rosuvastatin are not associated with clinically important adverse effects on adrenal and gonadal hormones. Randomized trials of lovastatin, simvastatin, and pravastatin did not find reductions in basal levels of cortisol or cortisol stimulated by adrenocorticotropic hormone (ACTH). Furthermore, studies of simvastatin 20 mg and 40 mg and pravastatin 40 mg compared with placebo found no differences in basal or stimulated testosterone, free testosterone, luteinizing hormone, or follicle-stimulating hormone. However, in one placebo-controlled trial of atorvastatin 80 mg, bioavailable testosterone was reduced by 10%, whereas total and free testosterone were unchanged. Although the clinical significance is unclear, statins do not cause hypogonadism or erectile dysfunction. A randomized trial of simvastatin 40 mg compared with placebo found no change in the menstrual cycle in women of reproductive age.

Randomized trials of lipid-lowering therapies that reduce LDL-C levels to < 50 mg/dL (< 1.293 mmol/L) have also found no adverse effects and/or no changes in steroid hormones. An analysis of the JUPITER trial, which randomly assigned participants to receive rosuvastatin 20 mg and placebo for a median of 2 years in 17,802 healthy people, and found no difference in adverse effects when comparing those who achieved or did not achieve LDL-C levels < 50 mg/dL (< 1.293 mmol/L). In the DESCARTES study of 901 patients on lipid-lowering therapy randomly assigned to receive evolocumab or placebo for 52 weeks, LDL-C levels were reduced to < 40 mg/dL (< 1.0344 mmol/L) in 87% of patients and to < 15 mg/dL (< 0.3879 mmol/L) in approximately 40%. Cortisol, ACTH, estradiol, and testosterone did not significantly change in either group.

Bile Acid Synthesis and Vitamin Absorption

Bile acids are needed for absorption of fat-soluble vitamins from the intestine. Studies have demonstrated that inhibition of hepatic cholesterol synthesis by a statin or by a statin in combination with an inhibitor of PCSK9 does not reduce bile acid synthesis. Furthermore, in situations of very low LDL-C, bile acid production is normal. Several studies have reported that low levels of LDL-C (due to lipid-lowering therapy) are not associated with lower levels of vitamin D or vitamin E. In the DESCARTES study, red cell vitamin E levels were stable, suggesting that tissue vitamin E levels were not reduced. Similarly, studies of statins have found normal red cell vitamin E levels. In addition, clinical vitamin E deficiency has not been observed in patients with very low LDL-C due to loss-of-function mutations in PCSK9.

Hemorrhagic Stroke

The concern that LDL-C reduction with statins would increase hemorrhagic stroke has been addressed in numerous clinical trials. In multiple randomized controlled trials of individuals with and those without a history of coronary artery disease, statins decreased ischemic stroke without an effect on hemorrhagic stroke. The effects of PCSK9 monoclonal antibodies in patients taking statins and achieving very low LDL-C levels were similar. However, in the randomized controlled SPARCL trial, high-dose atorvastatin (80 mg) compared with placebo significantly reduced ischemic stroke but increased hemorrhagic stroke in a small number of participants (55 in the atorvastatin group vs 33 in the placebo group) with a history of stroke or transient ischemic attack and no known history of coronary disease. Additional analysis of the data in SPARCL found that the small increase in hemorrhagic stroke occurred mainly in participants with a history of hemorrhagic stroke and was not related to baseline or recent LDL-C levels. A subsequent meta-analysis including randomized trials in which participants achieved LDL-C < 55 mg/dL (< 1.4223 mmol/L) found no increase in hemorrhagic stroke.

Cognitive Function

The hypothesis that low LDL-C causes cognitive decline has been investigated prospectively in the EBBINGHAUS study, which included a subgroup of patients in FOURIER (evolocumab). Cognition did not differ in patients randomized to receive evolocumab compared with placebo as assessed by the Cambridge Neuropsychological Test Automated Battery. Among patients assigned to receive evolocumab, cognitive function did not differ in those with low LDL-C (< 25 mg/dL [< 0.6465 mmol/L]) and higher levels of LDL-C. The caveat of this study is that median treatment was 19 months. Reassuringly, in FOURIER-OLE — a longer-term, open-label follow-up study that switched placebo-treated patients in the parent trial to evolocumab — there was no trend toward an increase in neurocognitive events during a maximum exposure to evolocumab of 8.4 years and mean LDL-C of 30 mg/dL (0.7758 mmol/L). There was no concurrent use of placebo during this follow-up trial.

Clinical Trials: Benefits of Achieving Very Low LDL-C

Overall, contemporary clinical trials of combination lipid-lowering therapy have generally shown greater CV risk reduction down to LDL-C levels < 25 mg/dL (< 0.6465 mmol/L), without any lower safety limit for LDL-C levels. Some of the strongest evidence comes from the FOURIER trial, a randomized, double-blind, placebo-controlled trial evaluating subcutaneous evolocumab (140 mg every 2 weeks or 420 mg monthly) vs placebo in 27,564 patients with atherosclerotic CV disease and a baseline LDL-C of 70 mg/dL (1.8102 mmol/L) on statin therapy. There was a 59% average reduction in LDL-C levels at 48 weeks with evolocumab compared with placebo — LDL-C reduction from baseline of 92 mg/dL to 30 mg/dL (2.379 to 0.7758 mmol/L).

Over a median follow-up of 2.2 years, evolocumab reduced the risk for major adverse cardiovascular events compared to placebo (9.8% vs 11.3%; HR, 0.85; 95% CI, 0.79 to 0.92; P < .001). These results were consistent for secondary endpoints and across key subgroups, without a significant difference between groups with respect to adverse events other than injection-site reactions (2.1% in the evolocumab group vs 1.6% in the placebo group). Of note, although the average LDL-C on treatment was 30 mg/dL (0.7758 mmol/L) in FOURIER, many patients had on-treatment LDL-C levels < 10 mg/dL (< 0.2586 mmol/L). These are valid LDL-C values because FOURIER performed ultracentrifugation on low LDL-C, giving confidence that these individuals truly had such low LDL-C levels (with subsequent validation of the Martin-Hopkins equation in this group). Moreover, there was no increase in adverse events — only additional CV benefit — in participants who achieved such low LDL-C. The ODYSSEY Outcomes trial results are generally consistent with FOURIER but were not as robust in demonstrating the safety of low LDL-C owing to dose adjustments in therapy that led to less achievement of very low LDL-C and more variable LDL-C levels.

The Way Forward

The several lines of evidence presented support the safety of very low levels of LDL-C (ie, < 25 mg/dL [< 0.6465 mmol/L]). Therefore, there is no compelling reason to reduce doses of lipid-lowering medications in adults with LDL-C < 25 mg/dL [< 0.6465 mmol/L]). Clinicians should reassure patients that such low levels are not only safe but beneficial. Lowering LDL-C for longer better protects patients from CV events such as myocardial infarction and stroke.

Follow Medscape on Facebook, X (formerly known as Twitter), Instagram, and YouTube