Blood testing is one of those areas where the gap between what most people assume they're getting and what they're actually getting is surprisingly wide. When your doctor says "your levels are normal," the unspoken question you almost never hear discussed is: normal where, in the fluid between your cells, or inside the cells themselves?
Let me explain why this distinction matters far more than most people realize.
When a lab processes your blood, it can measure nutrients and minerals in different compartments, and those compartments answer fundamentally different questions.
Serum (or plasma) testing measures what's circulating in the liquid portion of your blood, the extracellular space outside of your cells. This is what you get with the vast majority of standard blood panels. Think of it as checking what's currently in transit. It reflects short-term fluctuations: what you ate yesterday, whether you're dehydrated, whether you're fighting an infection. Your body aggressively maintains serum levels within tight homeostatic ranges, pulling minerals from bones and cells to keep the bloodstream looking "normal."
Serum magnesium, for example, represents less than 1% of your total body magnesium. [1]
Red blood cell (RBC) testing, by contrast, measures what's actually inside your cells, the intracellular environment where metabolic processes happen. Because red blood cells live for approximately 120 days, their mineral content reflects your nutritional status over the past three to four months. [2] It's the difference between checking your bank account balance (serum) and checking your actual net worth (RBC).
This is the same principle behind one of the most trusted tests in medicine: HbA1c, and one of the reasons it's part of Aniva's metabolic health panel. Rather than measuring blood sugar at one moment in time (fasting glucose, a serum-based snapshot), HbA1c measures glucose that has chemically bonded to hemoglobin inside red blood cells over their 120-day lifespan. [3] It is now widely accepted that HbA1c provides a far more reliable picture of metabolic health than any single fasting glucose reading. Fasting glucose can swing by up to 8.3% from day to day in the same person, while HbA1c variability is consistently below 3.5%. [4]
The medical world embraced this logic decades ago for blood sugar. It just hasn't applied it consistently to minerals and trace elements yet.
The magnesium example is the most well-documented and arguably the most consequential. A 2018 review in the journal Nutrients found that an estimated 60% of adults do not achieve adequate dietary magnesium intake, and roughly 45% of Americans may be deficient. [1] Yet this rarely shows up on standard blood work. The reason: the body pulls magnesium from bones and from inside cells to maintain serum concentrations within a narrow normal range (0.7–1.0 mmol/L). A person can have moderate to severe tissue-level depletion while their serum test reads perfectly fine.
A landmark paper in Magnesium Research explicitly warns that a normal serum magnesium level does not rule out deficiency. [5] A further review in Nutrients (2018) notes that the narrow serum range feeds the common clinical perception that magnesium levels "rarely fluctuate", which contributes to practitioners ordering the test infrequently, if at all. [2]
RBC magnesium testing can catch what serum testing misses, particularly in patients placed on long-term supplementation or depletion protocols of around three months or more, which aligns precisely with the red blood cell lifespan window. [2] This is why magnesium is one of the minerals Aniva tracks as part of its 140+ biomarker panel, because the number only means something if you know which compartment it came from.
Zinc tells a similar story. RBC zinc levels are less sensitive to short-term interference from meals, stress, inflammation, infection, or hormonal fluctuations than serum zinc. Research published in Endocrine Journal demonstrated that RBC zinc can even differentiate between different types of thyroid dysfunction, specifically distinguishing Graves' disease from transient thyrotoxicosis: a level of diagnostic specificity that serum zinc simply cannot provide. [6] In mild deficiency states, the body redistributes zinc from cells into the bloodstream to maintain serum appearance, masking the real deficit. [7]
For selenium, the picture depends on the chemical form and the timeline of exposure. Whole blood selenium incorporates the mineral bound within red blood cells and reflects longer-term selenium status, while serum selenium responds more quickly to recent dietary intake. [8] For chronic exposure assessments. Say, if you're supplementing with selenium for thyroid support (both selenium and thyroid markers are part of Aniva's core panel) — the RBC or whole blood measurement provides more clinically meaningful information.
One of the most compelling recent validations of this principle comes from the Omega-3 Index: the EPA+DHA content of red blood cell membranes, expressed as a percentage of total fatty acids. First proposed in 2004 as a cardiovascular risk factor, it has since been linked to outcomes across depression, cognitive decline, and all-cause mortality. [9] A 2013 study showed that an acute dose of fish oil raises plasma EPA+DHA by 47% within hours — but has no measurable effect on RBC levels. [10] The authors drew the direct parallel: just as HbA1c is unaffected by a single glucose spike, RBC omega-3 levels are unaffected by a single fish oil dose. Plasma levels, by contrast, can be gamed simply by taking a supplement the night before your blood draw.
A 2021 study analyzing over 25,000 individuals found that a low Omega-3 Index was directly associated with abnormal red blood cell distribution width. A marker independently linked to all-cause mortality. [11] This is an insight that serum fatty acid testing simply cannot provide, because it reflects membrane biology, not circulating lipids.
Curious what your own levels actually look like? Aniva tests 140+ biomarkers — including magnesium, zinc, selenium, Omega-3, HbA1c, CRP, and a full mineral panel. With results that tell you exactly which compartment was measured and what to do about it. The waitlist is free, and a full membership is just €199/year.
This is where the story takes a technical turn that directly affects anyone getting blood work done, including Aniva members.
When laboratories analyze trace elements in blood, they predominantly use a technique called ICP-MS (Inductively Coupled Plasma Mass Spectrometry). It is extremely sensitive and can detect elements at parts-per-billion concentrations. [12] But the critical variable isn't the machine. It's how the sample is prepared before it goes into the machine.
Serum samples are relatively simple to handle. They can be diluted with a mild acid solution (typically dilute nitric acid) and run through the spectrometer without much fuss. A 2023 study in BMC Chemistry demonstrated that 23 elements can be reliably measured in serum with simple 1:25 dilution in 0.5% nitric acid, no heating required. [13] The organic content is low, the matrix is straightforward.
Whole blood and RBC samples are a different animal entirely. Red blood cells have heavy organic matrices, proteins, lipids, hemoglobin, that must be broken down before the trace elements locked inside them can be measured. The standard approach for this is acid digestion: the sample is treated with concentrated nitric acid (HNO₃) and hydrogen peroxide (H₂O₂), then heated. Typically at 60°C or higher for 90 minutes or more, or subjected to microwave-assisted digestion at even higher temperatures and pressures. [12] This process essentially dissolves the cellular structure, liberating all elements into a measurable solution.
Here's the catch. When a lab uses acid digestion on a whole blood sample (which contains both serum and the cellular fraction), the heating and acid treatment rupture the red blood cells and release their intracellular contents into the surrounding liquid. The result is a single pooled measurement: a total value that combines extracellular and intracellular minerals into one number. It is not a serum value. It is not an RBC value. It's a hybrid that conflates two biologically distinct compartments.
Some laboratories, this practice has been documented in certain French lab protocols among others in Europe, use a unified acid digestion workflow for efficiency. A 2019 study from researchers at Hôpital Foch (Suresnes, France) and Université Paris Saclay described precisely this approach: a high-resolution ICP-MS method that measures 38 elements in human whole blood after microwave mineralization, using nitric acid and hydrogen peroxide for 25 minutes at high temperature. [14] The method is analytically excellent, validated, reproducible, suitable for detecting both essential and toxic elements. But the output is a whole-blood total, not a compartment-specific result.
A follow-up 2022 study by the same French research group applied this method to 20 autopsied subjects, measuring 39 trace elements across blood, hair, and organs. Confirming the approach is standard in French clinical and forensic toxicology practice. [15] The method is scientifically rigorous. But when these results are presented to patients or practitioners without explicit clarification that the number represents a digested whole-blood total rather than a pure serum or pure intracellular measurement, clinical interpretation becomes unreliable.
A Swiss validation study from 2020 (published in PLOS ONE) underscored this exact issue: the same single acid-based work-up was used for both serum and whole blood samples, but the resulting element concentrations were meaningfully different between the two matrices. [12] Iron, potassium, zinc, and magnesium in particular showed very different concentrations depending on whether the source was serum or intact whole blood — which makes biological sense, since the majority of these minerals are concentrated inside cells, not floating in serum.
A separate study in Biological Trace Element Research found that an initial attempt to digest whole blood with the same acid ratio used for serum left a flocculate, undissolved cellular material, proving that the two matrices require fundamentally different preparation protocols. [16] The whole blood matrix needed substantially more concentrated acid and hydrogen peroxide to fully dissolve, underscoring how different these sample types really are at the molecular level.
The reference ranges your lab uses may not match the sample preparation method. Standard serum reference ranges were established using centrifuged, non-hemolyzed serum. If a lab runs acid-digested whole blood but compares the result against serum reference ranges, the numbers will look artificially high for intracellularly concentrated elements like potassium, zinc, and magnesium. Conversely, elements that are predominantly extracellular may appear diluted. Most patients, and many practitioners, never think to ask how the sample was prepared.
If you're trying to assess whether you're chronically deficient in magnesium, a serum test is the wrong tool. If you're monitoring acute toxicity from a heavy metal exposure, whole blood or urine is appropriate. If you're tracking long-term nutritional optimization, which is exactly what Aniva members are doing, intracellular or RBC-level data is far more actionable. One sample type cannot answer every clinical question.
Hemolysis during collection can silently corrupt serum results. When a blood draw is traumatic, the needle is too small, the tourniquet is left on too long, the sample is shaken rather than gently inverted — red blood cells rupture and leak their contents into the serum. This is called hemolysis, and it artificially raises serum potassium, magnesium, iron, and zinc. A sample that should have shown normal serum levels now looks like a hybrid. Laboratories flag visibly hemolyzed samples (the serum turns pink), but mild hemolysis below the visible threshold can still shift results without triggering a flag.
Research shows that hemolyzed serum containing just 1 g/L of hemoglobin causes potassium to rise by 0.27–0.33 mmol/L, enough to push a borderline result into "high" territory. [17]
The "120-day window" isn't uniform. Red blood cells of different ages coexist in your bloodstream at all times. HbA1c research has confirmed that the measurement is weighted toward more recent weeks, approximately the past 30 days contribute disproportionately to the overall value. [3] The same applies to RBC mineral levels. This means a single RBC test captures a weighted rolling average, not a flat three-month mean. Two tests spaced eight to twelve weeks apart give far more information than one.
Each element has its own biology. A universal sample type applied to all elements is analytically convenient but clinically imprecise (this is why Aniva offers a dedicated heavy metals add-on with compartment-appropriate matrices for each element):
Serum iron, zinc, and cortisol all follow circadian patterns. A 1985 study in The American Journal of Clinical Nutrition measured serum zinc every 30 minutes across 24 hours in healthy adults and found a peak-to-trough difference of 19 µg/dL, with peak levels at 9:30 AM and the lowest point around 8 PM. [18] A 2018 NHANES analysis confirmed that time of blood draw significantly affects zinc assessment independently of diet or supplementation. [19]
For iron, a large study of 7,685 British men found clear diurnal shifts across 25 biochemical measurements: with potassium, hemoglobin, and hematocrit all higher in the morning and urea and creatinine higher in the afternoon. [20] A "normal" afternoon zinc result might have been flagged as "low" had the same person been drawn at 8 AM. Most labs standardize morning draws for this reason, but walk-in and direct-to-consumer services often do not enforce this.
Acute inflammation (even from a mild cold) drives serum zinc, iron, and selenium downward through a process called the acute-phase response — the body deliberately sequesters these minerals away from circulating pathogens. [2] When the inflammatory marker CRP is elevated, serum zinc and iron results can appear falsely low, potentially triggering unnecessary supplementation. [7] RBC mineral levels, by contrast, are far less affected by short-term inflammatory events — another reason they provide a more stable baseline for long-term health optimization. This is also why Aniva measures high-sensitivity CRP alongside minerals in the same draw — so your results can be interpreted in context, not in a vacuum.
Your blood test is only as good as what it measures, and how. Most panels check 20–30 markers. Aniva checks 140+ across minerals, nutrients, hormones, heart health, inflammation, thyroid, metabolic health, and more. One blood draw, one clear action plan, no medical jargon. Join the free waitlist and see the full biomarker list.
This is the insight that catches even experienced clinicians off guard. Proton pump inhibitors (PPIs), among the most prescribed drugs globally for acid reflux and heartburn — can cause clinically significant magnesium depletion by impairing intestinal absorption through the TRPM6 transporter. [21] The US FDA issued a safety warning in 2011 about this risk in long-term PPI users. [22]
What makes this especially insidious: PPI-induced hypomagnesemia is refractory to oral supplementation, high-dose magnesium supplements only partially restore levels, and only PPI withdrawal fully resolves the deficiency. [22] Yet because serum magnesium represents just 1% of body stores, many PPI users can be profoundly depleted at the cellular level while routine blood work shows nothing abnormal. The time to onset ranges from 14 days to 13 years with a mean of 5.5 years, making it easy to miss entirely. [23]
Similarly, loop and thiazide diuretics increase renal magnesium excretion, and the combination of a PPI plus a diuretic significantly amplifies the risk of hypomagnesemia. [24] If you're taking either of these medication classes, RBC magnesium testing is not optional, it's essential. Aniva's onboarding questionnaire captures your current medications precisely so these interactions are flagged in your personalized action plan.
Even before the lab touches your sample, the collection tube matters more than most people realize. Blood drawn into a tube containing potassium-EDTA (the anticoagulant used for hematology tests) will show dramatically different results than blood drawn into a plain serum tube. If tubes are filled in the wrong order during a draw, or if even trace amounts of EDTA carry over between tubes, the contamination can falsely elevate serum potassium while simultaneously lowering calcium, magnesium, and zinc. [25]
A UK study found that 28 out of 117 samples flagged for high potassium were actually contaminated with EDTA, and 20 of those had gone undetected by standard laboratory protocols. [26] Just 3.2% contamination with EDTA plasma caused an 11.9% increase in potassium and a 71% decrease in zinc in the same sample. [27]
There's also a subtler version of this: even the clotting process in a normal serum tube releases potassium from platelets, meaning serum potassium is inherently 0.36 ± 0.18 mmol/L higher than plasma potassium in the same blood draw. [17] In patients with elevated platelet counts (thrombocytosis), this effect can be large enough to mimic hyperkalemia.
Stop guessing. Start measuring what matters. Aniva's 140+ biomarker panel covers everything discussed in this article: minerals, Omega-3 status, metabolic markers, inflammation, thyroid, hormones, and more, with results designed for real people, not lab technicians. The waitlist is free. A full year of membership, including your blood draw, action plan, and doctor consultations, is €199.
Blood tests involve a bit more than just the number you get in a PDF. When you look at a blood test result, you're seeing the output of a specific analytical chain: which compartment was sampled, how it was prepared, what instrument analyzed it, and what reference range it was compared against. Each of those steps introduces assumptions, and when those assumptions don't match your clinical question, the number on the page can mislead.
For anyone serious about long-term health optimization rather than just screening for acute disease, intracellular and RBC testing provides the more meaningful signal for key minerals like magnesium, zinc, and selenium. Serum testing has its place, it's fast, cheap, and well-standardized, but it should be understood as a short-term snapshot of a tightly regulated extracellular environment, not a window into your actual cellular health.
At Aniva, our 140+ biomarker panel is designed with these distinctions in mind. We work with certified partner laboratories that use validated, compartment-appropriate methods — so when you see a result, you know exactly what it reflects and what to do about it. The waitlist is free, and a full year of membership, including your blood draw, clinician-reviewed action plan, and concierge support, is €199.
Ask your lab how they prepare your sample. Ask whether the result reflects serum, whole blood, or RBC levels. If they can't tell you clearly, that's a problem worth solving.
Medical disclaimer: This content is for informational purposes only and is not medical advice. Always discuss results with a qualified healthcare professional.
