The rest of this page explains why food has so little effect on this particular measurement, which parts of a lipid panel do move after eating, and what else is worth knowing before you book the draw.
Lipoprotein (a), often written Lp(a), is a cholesterol-carrying particle built on the same backbone as LDL, with an extra protein called apolipoprotein(a) attached to it. The blood test reports how much of that particle is circulating, either as a mass concentration in mg/dL or as a particle concentration in nmol/L. It is drawn from a standard venous blood sample, usually alongside a lipid panel, and the value is read against a laboratory reference range.
Because the biomarker page for this marker covers what the test measures, how it is run, the full range tables and how a single value is interpreted, this guide stays on the question you came here with. The short version is that this is a structural, largely genetic measurement rather than a metabolic snapshot of your morning.
A meal barely moves it because the particle is produced by the liver at a rate that is set mostly by your genes, not by the fat or sugar you ate an hour ago. Apolipoprotein(a) is encoded by the LPA gene, and the number of repeated protein segments (kringle repeats) in each person's copy of that gene strongly influences how much of the particle the liver releases into the blood. That production rate does not switch up or down in response to breakfast.
What does respond to breakfast is the transport of dietary fat. After eating, the gut packages fat into large triglyceride-rich particles that enter the circulation and are cleared over the following hours. Those particles are what make the postprandial lipid picture look different from the fasting one. Lp(a) sits outside that pathway. It is already in circulation, its clearance is slow, and a single meal does not meaningfully add to or subtract from the pool.
Research suggests that triglycerides are the lipid measure most affected by recent eating, while the remaining lipid measures change comparatively little in the hours after a meal 1. That is the practical reason most laboratories are comfortable running this marker on a non-fasting sample, and why a result drawn at two in the afternoon after lunch is generally treated the same way as one drawn at eight in the morning.
Triglycerides change the most, and everything else on a standard panel moves far less. Studies indicate that differences between fasting and non-fasting lipid measurements in general populations are modest for most parameters, with triglycerides showing the largest gap 2.
| Measure | Effect of a recent meal | Why |
|---|---|---|
| Triglycerides | Largest change of the panel | Dietary fat is carried in triglyceride-rich particles for several hours after eating |
| Total cholesterol | Minimal | Reflects cholesterol across all particles, which does not load up acutely from a meal |
| HDL cholesterol | Minimal | Not driven by the postprandial fat transport pathway |
| LDL cholesterol (calculated) | Small, indirect | Many labs derive it from a formula that includes triglycerides, so a raised triglyceride shifts the estimate |
| Non-HDL cholesterol | Minimal | Calculated by subtraction, so it is less sensitive to the triglyceride shift |
| Apolipoprotein B | Minimal | Counts atherogenic particles rather than the fat they are carrying |
| Lipoprotein (a) | Minimal | Production is largely genetically set and clearance is slow |
The column that matters for this page is the last row. If the only reason you were planning to skip breakfast was Lp(a), that fast is probably doing very little for the accuracy of the number. If the panel also includes triglycerides, or a calculated LDL cholesterol that depends on them, the picture is slightly different and is covered further down.
Many no longer do, though policies still vary by laboratory and by clinician. A 2024 review in Clinical Chemistry describes a worldwide shift toward non-fasting rather than fasting lipid profiles 3, and that shift is why the instruction you receive today may be different from the one you received years ago.
In practice, three things decide whether you are asked to fast. The first is the ordering clinician's preference and the clinical question behind the order. The second is the laboratory's own standing policy, which is printed on the requisition or sent with your appointment confirmation. The third is whether other fasting-sensitive tests are being drawn from the same sample, such as glucose or insulin, in which case the fast is being requested for those rather than for the lipid work.
The operational argument for non-fasting testing is straightforward. People spend most of their waking lives in a fed state, same-day testing is easier to complete, and fewer appointments get abandoned because someone forgot and ate. None of that changes the biology of this particular marker, which was stable either way. Follow whatever instruction your lab gives you, because a sample rejected on a technicality is worse than a sample drawn after a sandwich.
For general risk assessment, yes. Research suggests that non-fasting lipid profiles perform comparably to fasting profiles for cardiovascular risk prediction 4, which is a large part of why routine fasting requirements have been relaxed in many settings.
There is a useful nuance here. Whether fasting matters depends on the question being asked of the sample rather than on lipid testing as a category 5. A panel drawn to estimate ten-year cardiovascular risk is answering a different question from a panel drawn to investigate a markedly raised triglyceride level, and the second question may be better served by a standardised fasting draw.
For Lp(a) specifically, the fasting question is close to moot. The marker is not being used as a snapshot of recent metabolism. It is being used to describe a structural, largely inherited feature of your lipid biology, which is the same at breakfast as it is at midnight. The meta-analysis evidence on fasting versus non-fasting lipid testing in normal populations points in the same direction, with only modest differences for most measures 2.
No. The thresholds used to read an Lp(a) result are the same whether or not you ate, because the ranges were not built around a fasting state in the first place.
| Measure | Typical lab reference range | FitScript optimal range |
|---|---|---|
| Lipoprotein (a), mass units | 0 to 30 mg/dL | 0 to 14 mg/dL |
Typical lab reference ranges vary by laboratory, assay, age and sex, so read the range printed on your own report rather than a generic one. The optimal range shown here is FitScript's optimisation target, not a diagnostic cut-off, and a value outside it is not a diagnosis of anything.
One caution about units. If your report gives a value in nmol/L rather than mg/dL, the range beside it will be different, because the two units describe different things. A mass measurement and a particle-count measurement are not interchangeable, and comparing a number in one unit to a range in the other will produce a misleading conclusion. Check that the number and the range on your report share the same unit before you draw any inference from where the value sits.
A fast is still reasonable when something else on the requisition depends on it. The most common reasons are practical rather than related to Lp(a) itself.
The choice of fasting or non-fasting is best made against the specific clinical question rather than as a blanket rule 5. If you are unsure which applies to you, ask the ordering clinic before the appointment rather than guessing. Fasting unnecessarily is inconvenient; not fasting when the panel required it may mean a repeat draw.
It is considerably more stable than most of the lipid panel, though it is not perfectly fixed. Research on adult lipid clinic patients has examined how much Lp(a) varies between repeat measurements over time, and suggests that some within-person variation does occur rather than the value being entirely static 6.
That matters for how you read two results that do not match exactly. A modest difference between a result from last year and one from this year does not automatically mean something has changed in your physiology. Assay differences between laboratories, a change of units, normal biological variation and the timing of the draw relative to an illness can all contribute. The pattern across several results is usually more informative than the gap between any two of them.
Variability in lipid measurements over time is associated with cardiovascular risk 7, which is one reason clinicians tend to value a consistent testing routine. Using the same laboratory, the same unit and broadly the same conditions from one draw to the next removes noise that has nothing to do with your biology. That consistency is worth more than the choice between fasting and not fasting for this particular marker.
For most people, far less often than the rest of a lipid panel, because the value is largely set by inheritance rather than by day-to-day metabolism. Many clinicians therefore order it to establish a baseline and then revisit it only when the wider clinical picture changes or when a result needs confirming.
There are sensible reasons to repeat it. A single result sitting close to a decision threshold is worth confirming before anything is built on it. A result produced by a different laboratory or reported in a different unit is worth repeating under conditions that match your other records. A result drawn during an acute illness may be worth rechecking once you have recovered.
What does not make sense is repeating the test every few weeks in the hope of watching it move in response to short-term changes in routine. That is not how this marker behaves, and frequent retesting mostly generates noise. How often you retest is a decision for the clinician who ordered it, taking into account your family history, your other markers and what the first result showed.
Several things other than your last meal can nudge the number, which is why clinicians look at the context of a draw rather than the figure alone. Acute illness and significant inflammation can move it, because this particle behaves in part like an acute-phase protein. Pregnancy changes lipid physiology broadly. Kidney and thyroid conditions can influence circulating levels in people living with them, and clinicians factor that in when reading a report.
Technical factors matter at least as much. Different assays respond differently to the size variation of apolipoprotein(a) between individuals, so two laboratories can produce somewhat different figures from the same blood. Reporting in mg/dL versus nmol/L changes the number entirely. A change of laboratory between tests is one of the most common reasons a result looks like it has moved when the underlying biology has not.
None of these are reasons to dismiss a result. They are reasons to interpret it alongside when it was drawn, how you were at the time, and which laboratory ran it. If you were unwell in the days before your draw, say so when you discuss the result, because that single piece of context can change how the number is read.
In practical terms, yes. Fasting status has little bearing on this marker, but the unit on the report determines which reference range applies and whether two of your results can be compared at all.
A value in mg/dL describes the mass of lipoprotein (a) material per decilitre of blood. A value in nmol/L describes how many particles are present per litre. Because the size of apolipoprotein(a) varies substantially from person to person, a given mass can correspond to a different number of particles in two different people. That is why there is no single reliable conversion between the two units for an individual, and why laboratories publish separate ranges for each.
When you look at your own report, check three things before anything else. First, which unit the value is in. Second, which range the laboratory printed next to it in that same unit. Third, whether any earlier result you are comparing against used the same unit and the same laboratory. If the answer to the third question is no, treat the comparison with caution and raise it with your clinician rather than concluding that something has changed.
Lp(a) is most informative when it sits in a wider lipid and inflammatory picture, because it describes one specific particle rather than overall cardiovascular risk. The companion markers clinicians most often want on the same report include the following.
FitScript's membership includes 150+ biomarkers tested per year, with interpretation by Atlas, FitScript's AI health advisor, and any treatment decision made by a licensed provider after reviewing the member. The value of a wide panel here is context. A single number tells you about one particle; a panel tells you what that particle is sitting next to.
Follow the instruction your laboratory or clinician gave you, and make your conditions consistent from one draw to the next. For this marker specifically, that is most of the preparation.
Drink water normally, because being well hydrated makes the draw easier and avoids haemoconcentration effects. If a fast was requested, ask the clinic exactly how long it should be and whether plain water, black coffee or tea are permitted, since policies differ. Continue any prescribed medicines as your prescriber has directed unless that prescriber has told you otherwise; stopping something before a blood test is a decision for the person who prescribed it, not for the lab.
Try to keep the circumstances similar between tests. The same laboratory, a similar time of day and a similar general state of health all reduce the amount of difference between two results that has nothing to do with you. Mention recent illness, injury, surgery or pregnancy when you book or when you discuss the result, because those are exactly the contexts that change how a number should be read. And if you ate when you were not supposed to, tell the phlebotomist rather than staying quiet, so the report can be flagged accordingly.
Any Lp(a) result is worth discussing, and a result above the standard laboratory reference range is worth discussing promptly, particularly alongside a personal or family history of early cardiovascular events. A single value is an observation, not a diagnosis, and it cannot tell you on its own what will happen to you.
The useful way to approach it runs in three steps. The observation is where your number sits relative to the laboratory's range and, if you have one, your previous result in the same unit. The possible meaning is that a higher value may reflect an inherited tendency to carry more of this particular particle, which clinicians consider as one input among several when they assess cardiovascular risk. The next step is a conversation with a clinician who can place that number beside your other lipid markers, your blood pressure, your family history and your overall picture, and decide whether any further assessment is appropriate. If any therapy is considered, a licensed provider evaluates whether it is appropriate for you.
A value that is suboptimal rather than clearly outside the range still belongs in that conversation rather than in a search engine. Never ignore an abnormal result, and never stop or start anything on the basis of a single figure read on your own. This page is educational and is not medical advice. Discuss your results and any treatment decision with a clinician or licensed provider who knows your history.
For the wider question of whether to skip breakfast before a full lipid draw, see fasting before a lipid panel. If your result came back above the standard range and you want to understand what does and does not influence this marker, read how to lower lipoprotein a.
In most cases, no. The concentration of this particle is largely inherited and changes very little after a meal, so many laboratories accept a non-fasting sample. The deciding factor is your laboratory's own policy and whether other tests on the same requisition need a fast, so follow the instruction printed on your paperwork.
It is unlikely to move the number meaningfully. Dietary fat is carried by different, short-lived particles that mainly affect triglycerides. Lipoprotein (a) is produced by the liver at a rate set mostly by your genes, and a single meal does not change that rate.
Use the exact window your laboratory or clinician specifies, which is normally printed on the requisition or sent with your appointment confirmation. Ask the clinic directly if it is not stated, rather than assuming a length, because policies differ between laboratories.
Plain water is almost always fine and helps make the draw easier. Coffee and tea, with or without milk, are treated differently by different laboratories, so check with the clinic before your appointment if a fast was requested.
Far less than consistency does for this marker. Using the same laboratory, the same reporting unit and a similar time of day between tests removes variation that has nothing to do with your biology and makes two results easier to compare.
Usually not for this marker alone. A repeat may be sensible if the result sat close to a decision threshold, if it came from a different laboratory or unit, or if you were acutely unwell at the time. Your clinician decides whether a repeat draw is worthwhile.
One reports the mass of the material per decilitre of blood and the other reports how many particles are present per litre. They are not interchangeable, and each has its own reference range, so always compare your value against the range printed in the same unit.
Much less often than the rest of a lipid panel, because the value is largely inherited and does not move much over time. Many clinicians establish a baseline and revisit it only if the clinical picture changes or a result needs confirming. Your clinician sets the interval.
A single value is an observation, not a diagnosis. Bring it to a clinician, who can read it alongside your other lipid markers, your blood pressure and your family history, and decide whether any further assessment is appropriate. Do not start or stop anything on the basis of one number.