Nonfasting Lipids for All Patients?

Introduction

Change often takes time. In the medical community one of the more notable shifts in scientific thought and clinical practice has been the gradual acceptance of nonfasting lipid testing for routine screening of cardiovascular disease (CVD) risk. Internationally, the utility of nonfasting testing was first recognized in 2009 when the Danish Society for Clinical Biochemistry recommended its use over fasting profiles as the tests of choice for their national laboratories. Since then, several societies have approved of nonfasting lipid testing for routine screening (1–4). The United States has taken more time than the international community to translate scientific evidence regarding nonfasting lipid profiles into national guidelines. While the US Department of Veterans Affairs clinical practice guidelines preceded other societies by recommending nonfasting lipids in 2014, it was not until 2017 that the American Association of Clinical Endocrinologists and American Association of Endocrinology allowed for nonfasting testing (5). Finally, in 2018 the American College of Cardiology/American Heart Association (ACC/AHA) cholesterol guidelines modified previous 2013 recommendations for fasting, and allowed nonfasting testing for routine screening, reiterated again in the 2019 ACC/AHA prevention guidelines (6, 7). Table 1 summarizes key clinical guidelines and statements that recommend nonfasting for most routine evaluations. However, this acceptance is not universal; whereas about two-thirds of surveyed laboratories in Europe use nonfasting panels routinely (8), the 2017 Japanese and 2018 Korean guidelines prefer fasting tests. These differing practices lead us to the question: are nonfasting lipids suitable for all?

Table 1.

Key guideline and consensus recommendations on nonfasting lipid tests.

Guideline or statement	Year	CVD risk assessment or before starting lipid lowering therapy	During lipid lowering therapy	Nonfasting triglycerides
American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines (6, 7)	2018 and 2019	Nonfasting lipids are an acceptable alternative for fasting lipids for risk assessment in primary prevention and for assessment of baseline lipids in patients not yet on statin therapy.	LDL-C is the primary target.	Nonfasting triglycerides ≥175 mg/dL (≥2mmol/L) are abnormal and considered a risk enhancing factor. Fasting required for triglycerides ≥400 mg/dL (≥4.5 mmol/L).
European Atherosclerosis Society/European Federation for Laboratory Medicine/ European Sociey of Cardiology (1–3)	2016, 2018 and 2019	Fasting lipids are not routinely required	Fasting is not required if patients are on stable drug therapy	For triglycerides > 440 mg/dL ( >5mmol/L), fasting may be considered; refer to a specialist Nonfasting triglycerides ≥175 mg/dL (≥2 mmol/L) are elevated
American Association of Clinical Endocrinologists and American Association of Endocrinology (5)	2017	Nonfasting lipids are an acceptable alternative if fasting lipids are impractical	LDL-C is the primary target	Nonfasting triglycerides ≥150 mg/dL (≥1.7 mmol/L) are abnormal. Fasting required for management and treatment of hypertriglyceridemia
Canadian Hypertension Education Program Guidelines	2016	A fasting sample is no longer required; nonfasting is equally appropriate
Canadian Cardiovascular Society Dyslipidemia Guidelines (4)	2016	A nonfasting sample is considered an acceptable alternative to fasting	LDL-C is the primary target of therapy Non-HDL cholesterol or apoB are alternate treatment targets to LDL-C	For triglycerides >400 mg/dL (>4.5 mmol/l), repeat fasting triglyceride level
European Society of Cardiology Dyslipidemia Guidelines	2016	A nonfasting sample can be used in patients without severe hypertriglyceridemia or very low LDL-C	LDL-C is the primary treatment target Non-HDL cholesterol or apolipoprotein B are secondary targets	For general screening of hypertriglyceridemia, nonfasting triglycerides can be used in patients without severe hypertriglyceridemia
National Clinical Guideline Center (NICE) and Joint British Societies Guidelines	2014	A fasting sample is not needed	Consider an annual nonfasting non-HDL cholesterol	For triglycerides > 880 mg/dL (>20 mmol/L), refer to a specialist For triglycerides between 10 and 20 mmol/L (880 to 1770 mg/dL), repeat fasting
Veterans Affairs /Department of Defense Clinical Practice Guideline for the Management of Dyslipidemia for Cardiovascular Risk Reduction	2014	A nonfasting sample is recommended	Do not advocate treating to a target LDL-C or non-HDL-C for secondary prevention	For triglycerides >400 mg/dL (5.0 mmol/L), repeat fasting
American College of Cardiology/American Heart Association Guidelines	2013	A fasting sample is preferred (but not mandatory)	Fasting lipids to assess percent reduction in LDL cholesterol and adequate response to statin therapy	Increased nonfasting triglycerides ≥200 mg/dL (2.3 mmol/L) should be repeated fasting If ≥500 mg/dL (5.6 mmol/L), screen for secondary causes
Danish Society for Clinical Biochemistry	2009	A nonfasting sample is recommended.	NA	For TGs >350 mg/dL (>4 mmol/L), repeat fasting

Evidence for Nonfasting Tests

Universally the rationale behind using nonfasting testing is appealing physiologically. Except for a few hours in the morning, most of us spend the majority of our lives in the nonfasting state. Testing lipids and identifying risk while nonfasting therefore makes physiologic sense. In fact, the adequacy of nonfasting lipids for general screening of CVD risk has been verified by numerous large prospective studies over the past several decades (9). An evidence-based review of the published literature from >300 000 individuals found no diminution of lipid relationships for predicting incident events using nonfasting lipids (1). At least 3 large statin clinical trials have used nonfasting lipids (involving nearly 43 000 patients) (1). Concern that population level risk associations would not capture individual variability based on fasting status has been raised as a critique against adopting widespread nonfasting testing. Recent data published on 8270 participants from the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm (ASCOT-LLA) trial with prospective follow-up provided robust evidence addressing this concern (10). Both fasting and nonfasting lipids in the same individuals were measured 4 weeks apart with no intervention or advice given between the 2 visits. The association of baseline lipids with CVD events was similar, irrespective of fasting status, and importantly, results were similar by randomized allocation to statin versus placebo (10).

Studies also suggest that the postprandial state correlates with similar or even stronger risk associations of lipids with CVD, particularly for triglycerides (11). Genetic studies using Mendelian randomization have linked nonfasting triglycerides (and remnant cholesterol) to increased risk of CVD and mortality. In certain patients, including those with metabolic syndrome, diabetes mellitus, or specific genetic abnormalities, fasting can mask abnormalities in triglyceride metabolism that are captured by nonfasting measurements. Our recent study of a large primary prevention cohort found that, in addition to higher triglycerides, nonfasting lipids were associated with higher levels of large very lowdensity lipoprotein (VLDL), inclusive of chylomicrons, and medium sized VLDL cholesterol and particles, compared with fasting samples (12). Thus, nonfasting panels may help to identify residual lipid-related CVD risk with a proatherogenic milieu in patients, despite optimal guideline-based treatment.

One of the major reasons for slow adoption of nonfasting has been clinician concern regarding misclassification of individuals into a lower risk category when evaluating low density lipoprotein (LDL) cholesterol in nonfasting panels due to variability in triglycerides affecting LDL-C calculation via the Friedewald equation. However, we and others have found differences in routine lipids to be small: high-density lipoprotein [HDL] cholesterol change is negligible; slightly lower levels are seen [up to -8 mg/dL(0.21 mmol/L)] for nonfasting total cholesterol, LDL cholesterol, and non-HDL cholesterol compared with fasting; and modest changes [up to 26 mg/dL(0.3 mmol/L) higher] for triglycerides (7, 12). Risk algorithms use total cholesterol and HDL cholesterol (not triglycerides or LDL cholesterol), hence there is little impact of nonfasting on risk estimates using these methods. In addition, the ASCOT-LLA study found no significant misclassification that would adversely affect the decision for initiation of statin therapy, with high concordance (94.8%) between fasting and nonfasting lipids measured from the same individuals for classification into atherosclerotic CVD risk categories (10).

An alternative for clinicians concerned about using nonfasting LDL-C when titrating therapies for patients close to an absolute LDL-C target, is to use the Martin-Hopkins equation, a modification of the Friedewald equation with improved accuracy both when LDL-C is low and for nonfasting samples (7, 13). This is now the standard calculation method for LDL-C reporting by a large clinical laboratory provider. In many of the societal guidelines, a nonfasting non-HDL-C (or apolipoprotein B) may also be used to guide therapy. It is known that non-HDL-C level is a better risk marker than LDL-C, in particular in the setting of low LDL-C or when triglycerides are ≥200 mg/dL(2.3 mmol/L) (14). Treatment goals for non-HDL-C are 30 mg/dL higher than for LDL-C (fasting or nonfasting). Given the independent risk conferred by increased triglycerides and triglyceride-rich lipoproteins, the question of what constitutes high risk in terms of nonfasting triglycerides has also been studied. Most guidelines, including the latest US and European guidelines, define increased nonfasting triglycerides as ≥175 mg/dL (≥2 mmol/L), a cut point that has been identified and validated prospectively in a large study of US women (6, 15). The 2018 and 2019 AHA/ACC guidelines also consider fasting or nonfasting triglycerides greater than 175 mg/dL(2 mmol/L) as a risk enhancing factor that could prompt consideration for initiating or intensifying statin therapy (6, 7) To address potential ethnic variations, our analysis, that was enriched with African American participants, suggested that there were no significant race specific effects on lipoprotein levels and nonfasting status (12). When considering all of the above, the preponderance of evidence suggests that nonfasting screening would not only be acceptable, but in certain patients also provide more a tailored approach to CVD risk management.

Improved Healthcare Delivery and Experience with Nonfasting Tests

From a systems perspective, nonfasting screening allows for more efficient healthcare delivery and resultant patient and provider satisfaction. While there have been no studies to date assessing the cost-effectiveness of fasting versus nonfasting lipid testing, clinicians can relate to the experience of patients presenting for follow-up appointments without fasting. Management decisions are then deferred until follow-up tests and visits can be scheduled resulting in increased outpatient waiting times and potentially depriving others from access to care. Laboratory workflow can also suffer from an influx of early morning visits for fasting tests, decreasing system efficiency. It is not hard to surmise that decreased efficiency in multiple levels of the health care system leads to increased costs, burden on healthcare providers, and decreased patient satisfaction.

Recently the potential for fasting induced hypoglycemia in diabetic patients has been highlighted as an under-appreciated concern with more than 1 in 4 patients with diabetes experiencing a fasting-evoked en-route hypoglycemic event due to fasting for routine blood work (16). These hypoglycemic episodes add to patient morbidity avoided by adopting nonfasting screening.

Finally, concerns that patient consumption of a fatty meal prior to testing will result in markedly increased triglycerides and reduced test validity has been cited as a potential disadvantage for nonfasting samples. However, numerous studies have found that the increase in plasma triglycerides observed after habitual food intake is much less than that observed during a fat tolerance test, making this less of a concern for most patients (11, 17). In certain populations, simply advising patients to have a lighter meal or avoid fast food prior to their nonfasting blood draw would be sufficient counsel for patients.

When is Fasting Needed?

Robust and high-quality evidence supports the use of nonfasting lipid testing for the majority of patients. Fasting panels may be useful in selected patients prior to starting treatment that may itself result in or modify significant hypertriglyceridemia and in patients with genetic lipid disorders being followed for hypertriglyceridemia in lipid clinics. There is no consensus between guidelines as to the triglyceride cut-point that may prompt providers to order a repeat fasting panel at this time. While it may take some more time for universal acceptance of nonfasting screening, the benefits afforded by nonfasting tests such as improved triglyceride-related risk identification, avoidance of evoked hypoglycemia in patients with diabetes and more streamlined healthcare delivery are likely to apply to all healthcare systems and patients alike.