Diagnostic Use
Non-fasting Lipid Profiles
Non-fasted samples are acceptable in most cases of lipid assessment. Triglycerides can increase after a meal, but this is usually small (about 0.3 mmol/L) and therefore does not impact measures of LDL-, HDL- or non-HDL cholesterol except in extreme cases. In Auckland, there is a consensus from experts that non-fasting lipids can be used to assess absolute risk, using an algorithm such as PREDICT.
In most labs, including LabPLUS, LDL-cholesterol is calculated from other lipid measures.
Using the Sampson equation, LDL-results are valid as long as triglycerides are less than or equal to 9.0 mmol/L.
Above this level, a result will not be calculated.
Interpretation
Frequently asked questions:
What is the effect of a meal on lipid measurements?
There is little difference in most lipid parameters between fasting and non-fasting specimens after a typical meal (1-3). Total cholesterol and LDL fall by about 0.2 mmol/L and HDL by about 0.1 mmol/L, for up to 3-4h. The TC/HDL ratio does not change. These changes mostly reflect the dilution effect of water in the ingested food and are well within physiological variation of lipids typically seen in individual patients.
Why has fasting previously been recommended for lipid and CVD risk assessment?
Most large population studies and intervention trials (e.g. statin trials) have used fasting samples. However, there is no substantial evidence demonstrating fasting lipid levels to be superior to non-fasting levels for CVD risk prediction. The parameter used in current Framingham-based risk algorithms (TC/HDL) can be used equally well.
What evidence is there for the predictive value of non-fasting lipids on CVD risk?
Although evidence is less extensive than for fasting samples, a number of studies using non-fasting lipids (total cholesterol, HDL, TC/HDL ratio) have shown a strong association with CVD risk, and of similar magnitude to fasting results (about a 2-3 fold difference between subsets with highest and lowest TC/HDL) (4-6).
How is LDL calculated, and are LDL measurements valid in non-fasting samples?
In most laboratories LDL is calculated from the other lipid levels, rather than being measured directly (7, 8). Historically, the older Friedewald equation has been used for many years.
The Friedewald equation crudely estimates the cholesterol fraction in triglyceride-rich particles (VLDL, chylomicrons), and has been used, despite it’s problems, if the triglyceride result is less than 4.5 mmol/L. However, as triglyceride levels rise, even below 4.5 mmol/L, the calculated LDL using this formula becomes progressively biased low , especially in patients on aggressive LDL-lowering treatment. If the triglyceride is greater than 4.5 mmol/L, an LDL level cannot be calculated using the Friedewald formula and no result can be issued.
From 21/09/2021 LabPlus and the Chemical Pathologist Network of New Zealand has adopted a new formula, the Sampson formula, to calculate LDL. This equation is much better than the previous Friedewald equation at recognizing the non-linear contribution of cholesterol in triglyceride rich lipoproteins .
The Sampson calculation shows much less bias from triglyceride elevation, and is valid for triglycerides levels below 9.0 mmol/L. However, if the triglyceride is equal to or greater than 9.0 mmol/L, an LDL level still cannot be calculated.
What about triglycerides?
In healthy people triglycerides rise about 15-20%, or about 0.2-0.3 mmol/L (2,4), after a meal (for up to 6-8h)(1). Some patients have slow lipid particle clearance after food (so-called ‘post-prandial dyslipidaemia’). They typically have a high triglyceride and low HDL pattern even on fasting samples. However, non-fasting triglycerides may even better predictive value for future CVD events in such patients (9,10). Therefore an initial non-fasting sample is also reasonable. The disadvantage is that accurate ranges are less well documented than for fasting measurements, but non-fasting triglycerides above 2 mmol/L is suspicious for such ‘postprandial dyslipidaemia’ unless explained by some other factor, such as alcohol or oestrogens(11).
When, if ever, should fasting samples be taken?
The specific indications for a fasting are limited, especially as fasting glucose is also no longer considered the first line screening test for diabetes. The clearest reason is in patients with high triglycerides being monitored with lifestyle or drug intervention, where a f asting sample is better standardised. If fasting triglyceride levels fall below 4.5 mmol/L then calculated LDL can also be reported in such patients.
What is the value of non-HDL cholesterol?
The total of all atherogenic (apoB containing) particles can be expressed together as non-HDL cholesterol. This includes LDL as well as VLDL, IDL, and lipoprotein (a). Non-HDL compares favourably with LDL as a CVD risk marker, especially in patients with poor lipid particle clearance where the proportion of non-LDL particles in increase. It does not require the patient to be fasting. Non-HDL cholesterol can be easily calculated by the formula: Non-HDL cholesterol = total cholesterol – HDL cholesterol.
Although the modern emphasis is not on strict cholesterol treatment targets, but rather a risk-based approach, historically non-HDL treatment targets were usually about 0.8 mmol/L higher than LDL targets. In a non-fasting patient, a non-HDL level of 5.7 mmol/L or more may indicate genetic hypercholesterolemia that requires further evaluation or a secondary cause (13) .
Reference Intervals
All risk factors need to be taken into account when interpreting plasma lipid levels.
As a rough guide, the following levels are acceptable if no other major risk factors are present:
Adults
| total cholesterol |
<5.0 |
| HDL cholesterol |
>1.0 |
| LDL cholesterol |
<3.4 |
| Triglyceride (TG) |
<2.0 |
| total/HDL ratio |
<4.5 |
| Non-HDL cholesterol |
<4.2 |
Non-HDL cholesterol is a measure of cholesterol contained in all apoB containing particles, not just LDL. It includes upstream LDL precursors made by the liver, including VLDL, VLDL remnants/IDL, as well as LDL and lipoprotein (a). Chylomicrons originating from the small intestine have only small amounts of cholesterol in most patients and do not contribute significantly unless triglyceride levels are very high.
LDL are the smallest most atherogenic particles, and risk/ treatment guidelines have primarily focused on LDL, however these precursor particles are now also recognised as atherogenic. Epidemiological evidence for risk associated with non-HDL cholesterol is as strong or even stronger than LDL (Su 2019), although few randomised trials focus on non-HDL as a target.
In healthy patients with good lipid particle clearance cholesterol in upstream particles is usually present at low concentrations (0.8 mmol/L or less). However, their concentration increases in clinical states with slow clearance, e.g. diabetes, hypothyroidism, metabolic syndrome, drugs such as steroids or immunosuppressants, and some uncommon genetic conditions. Even if the LDL level appears reassuring, raised non-HDL cholesterol reflects increased concentrations of these other apoB-containing particles. While apoB levels can be directly measured, non-HDL cholesterol is a useful practical cost-effective alternative.
Although LDL remains the primary treatment target in international guidelines, focusing solely on LDL can be misleading in these patients. Most international guidelines therefore now consider non-HDL cholesterol as a secondary treatment target, with levels 0.8 mmol/L higher than the LDL target, e.g. if target LDL is <1.4mmol/L in high risk patients, the target non-HDL is <2.2 mmol/L.
Su X, Kong Y, Peng D (2019). Evidence for changing lipid management strategy to focus on non-high density lipoprotein cholesterol. Lipids in Health and Disease 18, number 134.
Visseren FLJ, Mach, F, et al (2021). ESC Guidelines on cardiovascular disease prevention in clinical practice: Developed by the Task Force for cardiovascular disease prevention in clinical practice with representatives of the European Society of Cardiology and 12 medical societies With the special contribution of the European Association of Preventive Cardiology (EAPC). Euro Heart J 42(34):3227-37
Williams E, Round T, Jones NR (2024). Cardiovascular disease – risk assessment and reduction: NICE 2023 update for GPs. Br J Gen Pract 74(748):523
Patients <19 years old
| Age |
Total cholesterol |
HDL cholesterol |
Triglyceride |
| 0 – 14 days |
1.2 – 3.2
|
0.4 – 1.1
|
0.9 – 3.0
|
| 15 days – 1 year |
1.6 – 6.2
|
0.3 – 1.9
|
0.6 – 3.0
|
| 1 to 19 years |
2.9 – 5.4
|
0.8 – 1.9
|
0.5 – 2.3
|
Units : mmol/L
Conversion factors :
Cholesterol (total, HDL and LDL): mg/100 mL x 0.026 = mmol/L
mmol/L x 38.6 = mg/100 mL
Triglyceride: mg/100 mL x 0.0113 = mmol/L
mmol/L x 88.5 = mg/100 mL
Test Method
Total Cholesterol
Principle : Enzymatic/ Colorimetric
Reagents: Roche CHOL2 kit
Analyser: Cobas c702
HDL Cholesterol
Principle : Enzymatic/ Colorimetric
Reagents: Roche HDLC4 kit
Analyser: Cobas c702
Chol/HDL Chol Ratio
Principle : This is calculated as follows:
Chol/HDL Chol Ratio = total cholesterol / HDL-cholesterol.
LDL Cholesterol
Principle : This is calculated using the Sampson equation as follows:
LDL Cholesterol = 1.055 x total cholesterol - 1.03 x HDL-cholesterol - [(triglycerides/3.735) + triglycerides((total cholesterol - HDL-cholesterol)/24.18) ((triglycerides2)/79.38] - 0.244.
This calculation is valid when serum triglycerides are less than 9.0 mmol/L.
Triglyceride
Principle : Enzymatic/ Colorimetric
Reagents: Roche TRIGL kit
Analyser: Cobas c702
Non-HDL Cholesterol
This is not reported, but can easily be calculated as follows:
Non-HDL Cholesterol = total cholesterol - HDL-cholesterol.
Uncertainty of Measurement
Uncertainty of Measurement:
total cholesterol 4%
HDL-cholesterol 6%
Triglyceride 5%