Many US lab reports print the sample type inside the test name: "Potassium, Serum", "Glucose, Plasma", "Protein, Total, Serum". Nordic reports use a prefix instead, S- for serum and P- for plasma. For most lines on a report that label changes nothing. For about six results it explains why a number sits a little high, why it differs from last year's result at another lab, or why a doctor wants a repeat before acting on it.
Serum is plasma that has been allowed to clot
What serum and plasma are
Both serum and plasma are the straw-yellow liquid left once a lab spins blood in a centrifuge and the cells sink to the bottom. The difference is what happened in the tube first. Plasma comes from blood mixed with an anticoagulant, a chemical that stops clotting, so it keeps fibrinogen and the other clotting proteins. Serum comes from blood left to clot, usually for about 30 minutes in a tube with a clot activator, so those clotting proteins are used up in the clot 1. Serum, in other words, is plasma minus the clotting proteins, plus whatever the cells released while the clot formed. Plasma can be spun straight away and the same tube of blood yields roughly 15 to 20% more plasma than serum, which is why hospitals that need fast results lean on plasma 1.
The cap color on the tube tells you which one you gave. Colors follow a common convention, though brands vary slightly. Phlebotomists fill tubes in a set order of draw so that one tube's additive does not carry over on the needle into the next 3.
| Cap color | Additive | Gives | Typical tests |
|---|---|---|---|
| Red | None, or a clot activator | Serum | Chemistry, drug levels, antibodies |
| Gold or red-grey (serum separator tube) | Clot activator and a gel barrier | Serum | Metabolic panel, lipids, thyroid, vitamins |
| Green | Lithium or sodium heparin | Plasma | Fast chemistry, potassium repeats |
| Lavender or purple | EDTA | Whole blood or EDTA plasma | CBC, A1C, blood typing |
| Light blue | Sodium citrate, filled to 9 parts blood to 1 part citrate | Plasma | PT/INR, aPTT, fibrinogen, D-dimer |
| Grey | Sodium fluoride and potassium oxalate | Plasma | Glucose, lactate |
What clotting changes in the liquid
Clotting takes something out and puts something in. It takes out fibrinogen, the protein that turns into the fiber mesh of a clot. Fibrinogen normally runs at about 2 to 4 g/L, so plasma total protein reads a few g/L higher than serum total protein from the same blood 1. Clotting puts in what platelets (the small cell fragments that start a clot) and other blood cells release as they activate: potassium, phosphate and LDH (lactate dehydrogenase, an enzyme found inside most cells) all read higher in serum than in plasma 13. White cells also spill their DNA into the clotting tube, so serum carries several times more cell-free DNA than plasma 6.
Plasma carries its anticoagulant into the test, and that causes its own problems. A lithium heparin tube cannot measure a lithium drug level, heparin can block PCR-based tests, and EDTA works by binding calcium, so EDTA plasma is wrong for calcium 13. Labs choose the tube for each test with these effects in mind, and they set each test's reference range on the sample type they use.
Which results change with the tube, and which don't
| Result | Serum vs plasma | Why | What it means for you |
|---|---|---|---|
| Potassium | Serum higher, usually by a few tenths of a mmol/L; more with a high platelet count | Platelets release potassium as the clot forms 12 | A borderline high serum potassium is often repeated, sometimes as heparin plasma |
| Phosphate | Serum slightly higher | Released by cells during clotting 1 | Compare serum with serum, plasma with plasma |
| LDH | Serum higher | Released by platelets and cells during clotting 1 | Compare like with like; hemolysis raises it further |
| Total protein | Plasma higher by roughly the fibrinogen, 2 to 4 g/L | Fibrinogen is used up in the clot 1 | Read against the range printed for that sample type |
| Glucose | Falls in any unspun tube; serum sits about 30 minutes while it clots | Blood cells keep burning sugar 4 | A delayed sample can read low |
| PT/INR, aPTT, fibrinogen | Citrate plasma only | Serum has no clotting proteins left | No serum version exists |
| Sodium, chloride, creatinine, urea, cholesterol, ALT, most hormones and antibodies | Practically the same | Not stored in platelets, not used up by clotting 1 | Compare freely, ideally at the same lab |
Results where the tube shows
Potassium: the result where it matters most
Serum potassium runs higher than plasma potassium because platelets release potassium while the blood clots in the tube 1. In most people the gap is a few tenths of a mmol/L, small enough to sit inside the reference range. It grows when the platelet count is very high, because more platelets release more potassium in the tube 2. Very high white cell counts, as in some leukemias, can also push serum potassium up, mainly when a sample is delayed or handled roughly, though the 2008 review's own data found no consistent white cell effect 2. When the gap is big enough to make a normal person look hyperkalemic (high in potassium), it has a name, pseudohyperkalemia, defined in a 2008 review as "a rise in serum potassium concentration with concurrently normal plasma potassium concentration" 2.
Clotting is one of several ways a tube can raise potassium. Pumping a fist during the draw, a tourniquet left on too long, blood that sits unspun for hours or gets chilled before spinning, and red cells breaking open (hemolysis) all push it up. So does EDTA from a lavender tube carried over into a chemistry tube, which reads as high potassium and low calcium together, a pairing labs recognize as a sign of contamination 3. Hemolysis has its own page, the hemolyzed blood sample guide.
Red cells hold far more potassium, LDH and AST (two enzymes found inside cells) than the liquid around them, so those results can read falsely high.
When a high serum potassium surprises a clinician, the usual moves are to look at the platelet and white cell counts on the same day, ask how the sample was drawn and handled, and repeat it, often as heparin plasma, which skips the clotting step 12. If serum reads high and plasma from the same draw reads normal, that points to a tube effect, and your clinician reads it as such. For how potassium fits with sodium, chloride and CO2, see electrolytes explained.
Read these beside a high potassium
Glucose: why the tube and the clock matter
Blood cells keep burning glucose after the draw until they are separated from the liquid. In unspun blood, glucose falls by about 5 to 7% an hour. Fluoride, the additive in grey tubes, does almost nothing in the first hour and can take up to 4 hours to stop the fall 4. A serum tube adds its 30-minute clotting wait on top. For diagnosing diabetes, the 2023 laboratory guideline from the American Diabetes Association and the Association for Diagnostics and Laboratory Medicine (ADLM) recommends venous plasma in a tube with a fast-acting glycolysis inhibitor such as citrate buffer, or a tube placed straight into an ice-water slurry and spun within 15 to 30 minutes 4.
Fingerstick meters read whole blood, which runs about 11% lower than plasma, so the International Federation of Clinical Chemistry (IFCC) recommends that all glucose results be reported as plasma values, converting whole-blood readings with a factor of 1.11 5. Most meters now do that conversion for you, but a fingerstick and a lab result still rarely match exactly, because meters carry a wider margin of error than a lab analyzer; see blood sugar explained and CGMs for people without diabetes. A1C is measured on whole blood from a lavender EDTA tube and none of the serum-or-plasma timing applies to it.
The glucose results on your report
Can you compare a serum result with a plasma result?
For most markers, yes. Sodium, chloride, creatinine, cholesterol, liver enzymes other than LDH, and most hormone and antibody tests agree between serum and heparin plasma when samples are spun promptly 1. For potassium, phosphate, LDH, total protein and glucose, compare serum with serum and plasma with plasma, or allow for the known shift. Each lab sets its reference interval on the sample type it uses, so the range printed beside your number already accounts for it. A plasma potassium range can stop a few tenths lower than a serum range, and reading a plasma result against a serum range from the internet can make a borderline value look normal. More on why ranges differ in reference ranges explained. For tracking a marker over months, the cleanest trend comes from the same lab and the same sample type; see trends beat single readings.
Home kits, cancer tests and other special cases
- Many finger-prick kits collect blood into small gel tubes that clot into serum and travel by post unspun. The days in the mail matter most for potassium and glucose, because cells left in the tube keep releasing potassium and using up glucose. See the finger-prick blood test guide.
- Cell-free DNA and multi-cancer early detection tests such as Galleri use plasma in special stabilizing tubes, because white cells spill their own DNA into serum while it clots 6.
- The gel in serum separator tubes can absorb some drugs, including phenytoin, phenobarbital, carbamazepine and lidocaine, lowering measured levels on stored samples, so drug-level tests may call for a plain tube 3.
- Clotting tests need a light blue citrate tube filled to the line. An underfilled tube changes the blood-to-citrate ratio and skews the result; see clotting tests explained.
Reading your own report
- Find the sample type on each line: "Serum", "Plasma", or an S- or P- prefix. Some labs put it in the test name, others in a specimen column; see how to read Quest results and how to read Labcorp results.
- Check the range printed beside the number, which the lab set for that sample type, before checking any range from elsewhere.
- If a surprise result is potassium, phosphate, LDH, total protein or glucose, ask whether a repeat, or a different tube, would change it before reading it as a change in your body.

Questions
Is serum the same as plasma?
No. Both are the liquid part of blood, but plasma comes from blood kept liquid with an anticoagulant and still holds fibrinogen and the other clotting proteins, while serum comes from blood that clotted first, so those proteins are gone. For most lab results the two give practically the same number; for potassium, phosphate, LDH, total protein and glucose they differ.
Why is serum potassium higher than plasma potassium?
Platelets release potassium while blood clots in a serum tube, so serum potassium usually reads a few tenths of a mmol/L above plasma potassium from the same draw. The gap widens with a very high platelet count, and very high white cell counts can do the same, mainly when a sample is delayed. A serum value more than 0.4 mmol/L above plasma is called pseudohyperkalemia, a tube effect a clinician confirms with a repeat.
Which is more accurate, serum or plasma?
Neither is more accurate across the board. Plasma is closer to the circulating level for potassium, phosphate and LDH and is the only option for clotting tests; serum avoids anticoagulant interference and is the validated sample for many hormone and antibody tests. Accuracy depends on the lab using the sample type its test and reference range were set up for.
Can I compare a serum result with an old plasma result?
For sodium, creatinine, cholesterol and most hormones, yes. For potassium, phosphate, LDH, total protein and glucose, compare like with like or allow for the known shift, and judge each result against the range printed on its own report.
Is donated plasma the same as plasma in a lab test?
It is the same liquid. A donation center separates plasma from your blood with an anticoagulant and returns the cells; a lab tube does the same in miniature with a few milliliters, then measures what is in it.
Citations
- World Health Organization. Use of anticoagulants in diagnostic laboratory investigations. WHO/DIL/LAB/99.1 Rev.2. Geneva: WHO; 2002.
- Sevastos N, Theodossiades G, Archimandritis AJ. Pseudohyperkalemia in serum: a new insight into an old phenomenon. Clin Med Res 2008;6(1):30-32.
- Bowen RAR, Remaley AT. Interferences from blood collection tube components on clinical chemistry assays. Biochem Med 2014;24(1):31-44.
- Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2023;69(8):808-868.
- D'Orazio P, Burnett RW, Fogh-Andersen N, et al. Approved IFCC recommendation on reporting results for blood glucose. Clin Chem 2005;51(9):1573-1576.
- Lui YYN, Chik KW, Chiu RWK, et al. Predominant hematopoietic origin of cell-free DNA in plasma and serum after sex-mismatched bone marrow transplantation. Clin Chem 2002;48(3):421-427.
Educational context only, not medical advice or a diagnosis. Always discuss your results with a clinician.
