Can iron be normal on a blood test and still limit training? What the reference range means

A reference page on what the edges of a blood test range were built to mark, why ferritin is the hardest value to read against them, and what to ask before a result inside the lines counts as enough for someone who trains.

Yes, it can. A ferritin inside the printed range means your value resembles the values of the crowd the laboratory used as its reference, and nobody in that crowd was selected for training hard around a full working week. The laboratory compared you with other people. Nobody compared you with your week.

This is part three of the Metabolic series. Part two asked whether creatine works for everyone and ended on the headroom a person brings to a molecule; this page is about a blood value that can hide how little headroom is left. The Substack essay tells the same argument in a shorter, more personal way. Here it's unfolded study by study, with each study's limits named and every source listed at the end.

Key points

A reference range describes the middle 95 percent of a group presumed healthy. It carries no information about what a body in hard training needs.
WHO counts a ferritin below 15 µg/L as iron deficiency in a healthy adult, and bone marrow comparisons put 15 roughly where stainable iron runs out.
Ferritin climbs with inflammation. WHO's line for an inflamed adult is 70, and a pooled analysis of 32 studies found inflammation lifted ferritin by about 30 percent.
In a small study, one hard hour of running left the inflammation marker CRP at 2.6 times its starting value a day later, and it raised hepcidin, the hormone that limits how much iron the gut lets through, for several hours afterwards.
Pushing iron into stores that were adequate adds risk.
In a 2018 review, every trial that found a performance gain from iron had enrolled people with a ferritin of 20 or below.
Three questions make a ferritin result readable: was inflammation measured beside it, which range applies, and when in the training week the blood was drawn.

This is an educational and strategic perspective, not personal medical advice.

The views are the author's own and not statements by Atlas Cove Lda.

What a reference range is built from

A laboratory that prints a range has, at some point, measured a large group of people it had reason to think were healthy, sorted their results, trimmed a thin slice off each end until 95 percent remained, and printed the two cut points as your lower and upper limit. That's a description of a crowd. It's a good one, and it answers exactly one question, which is how unusual your number is among those people.

Whether the number is enough for your life is a different question, and the range was never designed to answer it. The crowd behind it wasn't chosen for training hours, for a hard week at work, or for anything else about how you spend your days. An inside-the-lines result tells you that you look typical, and typical is a statement about other people.

I learned this late. For years I gave blood regularly, so my iron got checked more often than most people's, and it always came back fine. Later I had a couple of years of winters and springs that went badly, with heavy migraines, a flu-like ache and a tiredness that could take a whole day, and every panel from that stretch came back unremarkable, iron included. I treated each report as closed the moment it came back inside the lines. I never asked what the lines were drawn against.

If you've ever walked out with a clean printout and exactly the tiredness you walked in with, my co-founder Lisa wrote about that feeling in your labs are fine, your life is not. This page is the narrower, mechanical version, using one value where the gap between normal and enough is unusually easy to see.

Why the lower edge of the ferritin range sits at empty

Ferritin is the protein that stores iron inside your cells, and a little of it leaks into the blood in proportion to what's stored, which is why it became the standard marker for iron stores. WHO's guideline puts the threshold for iron deficiency in an apparently healthy adult at a ferritin below 15 micrograms per litre (WHO, 2020).

WHO's own nutrition department had already checked whether 15 was right. Garcia-Casal and colleagues gathered the studies that had set ferritin beside a bone marrow sample, the only direct way to see stored iron, and looked at healthy people. Across nine studies with 390 participants whose marrow showed no stainable iron at all, mean ferritin was 15.1. Across three studies with 151 participants whose marrow showed some iron, the mean was 70.4 (Garcia-Casal et al., 2018). All of those studies were observational, and the second group is small, but the direction is plain.

Fifteen is where the marrow runs dry. A line drawn at empty is a line for absence, and a result of 20 or 25 clears it while still sitting nearer the empty group's mean than the mean of the group with iron visible. I find that uncomfortable, mostly because I'd have read a value like that as fine without thinking twice.

Sports medicine draws its lines higher. The Swiss consensus from 2015, written by the country's sports physicians, calls values under 15 empty, calls 15 to 30 low stores, and treats 30 as the sensible cut-off for healthy athletes older than fifteen; the same paper names 50 as the value its authors want to see before altitude training, where the demand for iron rises (Clénin et al., 2015). Those figures are one group's consensus for its own athletes. I quote them to show that two sets of professionals read the same value against different floors, and a printed report doesn't tell you which floor it used.

A donor check that clears empty stores

Blood donation is where my own habit came apart, so it gets its own section. Before a service takes your blood it tests whether you can spare it, and in the United States the test is a finger-prick haemoglobin with a floor of 12.5 g/dL (Cable et al., 2011). Pass it and you donate.

Cable and colleagues enrolled 2,425 American donors in the RISE study, every one of whom had just passed that finger-prick and been accepted. Fifteen percent had absent iron stores, defined as a ferritin under 12. Among frequent donors it reached 16 percent in men and 27 percent in women. The odds climbed with every donation: against first-time donors, people with four or fewer donations in the previous two years had 5.3 times the odds of absent stores, and people with seven to nine donations had 13.5 times the odds.

How can a donor pass on empty? Stores go first. Haemoglobin holds up until late in the process, long after the reserves have started to run down, which is why there's a whole research literature on athletes who are iron-deficient while their haemoglobin still looks perfectly normal (Rubeor et al., 2018). A haemoglobin check answers whether you can spare a donation today. It was never built to say what's left in the cupboard.

For years I took "you can donate" to mean "your iron is fine". Worse, I'd assumed that if donating did anything to my iron, it protected me from carrying too much. Every donor in that study cleared the same kind of check I cleared.

Why inflammation lifts ferritin, and by how much

Ferritin has a second job. It's an acute phase protein, one of the proteins whose level in the blood climbs whenever the body is dealing with an infection, an injury or a hard stretch of stress, so it rises during inflammation whether or not a single extra atom of iron has been stored. A store that's genuinely small can therefore post a respectable number.

WHO builds this into its thresholds. Its guideline says markers of inflammation should be assessed together with ferritin, and it sets the deficiency line for an adult with infection or inflammation at a ferritin below 70, against 15 for an adult who is well (WHO, 2020). Which line applies depends on a value most standard panels leave out.

How big is the effect? Thurnham and colleagues pooled 32 studies of apparently healthy people who had ferritin measured alongside two inflammation markers, C-reactive protein and alpha-1-acid glycoprotein. Overall, inflammation raised ferritin by about 30 percent, and ignoring it meant iron deficiency was underestimated by 14 percent (Thurnham et al., 2010). And those were people described as apparently healthy.

Hard training produces a short inflammation of its own. Peeling's group put eight moderately trained runners, six men and two women, through an hour of running: fifteen minutes at 75 to 80 percent of peak heart rate, then forty-five at 85 to 90 percent, compared in a crossover with a trial of sitting at rest. Interleukin-6, an early signal of inflammation, was 6.9 times its resting value the moment the run ended. C-reactive protein, or CRP, the marker a doctor is most likely to add to a panel, still stood at 2.6 times its starting value a full day on (Peeling et al., 2009).

Blood drawn the morning after a hard session comes from a body still reporting that session. I should be careful here. Peeling's study didn't measure ferritin on the next morning; putting its CRP finding together with WHO's inflammation rule is my own step, and it's an inference. Still, it's the inference I'd want whoever reads my ferritin to have made.

How hepcidin decides what gets through the gut wall

Your body can't get rid of iron deliberately. There's no organ that excretes a surplus, so every bit of control has to happen on the way in, at the gut.

The controller is hepcidin, a small peptide hormone the liver makes. Iron leaves the cells lining your intestine through a transport protein called ferroportin. Nemeth and colleagues showed that hepcidin binds to ferroportin, and that the bound protein is then taken inside the cell and broken down, so less iron is exported into the blood regardless of how much arrived in the meal (Nemeth et al., 2004). Plenty of iron makes the liver release more hepcidin, and so does inflammation. Too little of it and iron accumulates in tissues. Too much and iron in the blood runs low while the stores sit where they were. That work was done in cultured cells, so it describes the mechanism rather than a measured effect in people.

In Peeling's runners, hepcidin measured in urine was 1.7 to 3.1 times its pre-run level at three, six and 24 hours afterwards, and three hours after the run it was three times the value at the same hour on the rest day. It rose in every one of the eight, though by very different amounts (Peeling et al., 2009). For those hours the entrance is narrower. A meal meant to replace iron that lands inside that stretch gets less of its iron through.

McCormick and colleagues then asked whether timing matters. Sixteen runners, ten men and six women, all with a ferritin below 50, ran for 90 minutes at 65 percent of the speed at which they reached their VO2 max, in a morning session and, on another occasion, an afternoon one, and ate a standard meal carrying labelled iron 30 minutes later; fourteen days after each meal the researchers measured how much of the label had turned up in red blood cells (McCormick et al., 2019). Hepcidin was higher three hours after exercise, as expected. It also drifts upward over an ordinary day by itself: in their afternoon trial, hepcidin at rest had climbed by 0.55 nM across the day.

The absorption result is the odd part. The breakfast eaten straight after the morning run was absorbed better than breakfast on a rested morning, and better than a dinner later on the day of the morning run. The authors read that as a short window straight after exercise, before the hepcidin rise takes hold. I like this finding because it complicates a tidy story, and I'm wary of it for the same reason: sixteen people, all with fairly low stores, one run at each time of day. I draw no meal-timing rule from it. What it does show is that the hour you eat relative to the hour you train is a real variable, and no blood report records it.

Why low stores switch the hepcidin rise off

If hard training closes the entrance, the obvious worry is a spiral: train, absorb less, store less, train again. The body guards against that too.

Five years on, the same group combined 54 athletes from five earlier studies of theirs and split them by resting ferritin into four groups: under 30, 30 to 50, 50 to 100, and over 100. After exercise, hepcidin rose significantly in the three higher groups. Below 30 there was no significant rise (Peeling et al., 2014). The authors put it plainly: low stores suppressed the post-exercise rise, seemingly overriding the inflammatory signal. When the stores are low, the body's need for iron seems to outrank the inflammation.

One detail in that paper deserves more attention than it usually gets. The groups split by sex almost as sharply as by ferritin: eight of the twelve athletes under 30 were women, and eighteen of the twenty above 100 were men. The ferritin groups and the sex groups overlap heavily, so the study can't cleanly separate the two. I mention it as a description of who was in the groups, nothing more.

Why more iron is the wrong reflex when stores are adequate

Once you've read that far, the instinct is predictable. Stores might be low, absorption is throttled after training, so take more iron. I had the same reflex.

There's an upper edge, though, and WHO draws it: for adults who are otherwise well, a ferritin above 150 in women and above 200 in men marks a risk of iron overload (WHO, 2020). Remember that nothing excretes the surplus. The Swiss consensus says that long-term daily iron taken by mouth, or given by infusion, when ferritin is normal or high, makes no sense and may do harm (Clénin et al., 2015). Someone raising a value that was never low has taken on a risk to change a number.

And the benefit is less secure than people assume, which is what persuades me most. In Klingshirn's trial, eighteen women who ran distance, all with a ferritin under 20, took iron or a placebo for eight weeks. Ferritin ended at 23.4 with iron and 15.7 with placebo. Time to exhaustion improved by 25.5 percent with iron and 22.2 percent with placebo, a difference that was nowhere near significant (Klingshirn et al., 1992). Rubeor and colleagues reviewed twelve trials with 283 participants in 2018: six found a performance gain and six didn't, and all six that did had admitted only people whose ferritin sat at 20 or lower (Rubeor et al., 2018). Near empty, iron can matter a lot. Above that, I haven't seen good evidence that raising the number changes anything else.

Women who menstruate start from a different place, and the numbers show it. A 2019 narrative review found iron deficiency in roughly 15 to 35 percent of the women in the athlete groups studied, compared with about 5 to 11 percent of the men (Sim et al., 2019). It describes cohorts, and it marks the end of what I can usefully say. The questions below are the same for everyone, and when a woman trains hard, her own physician is the person to answer them.

Three questions to bring to your next blood report

None of this needs anything bought. It needs three questions, asked of whoever reads your results, and it needs no target from me, because the point of the questions is that the person with your whole picture in front of them decides what your number means.

Was an inflammation marker such as CRP measured together with ferritin, and what did it say about the state your body was in that morning?
Is the report measured against the general population range or a range for people in training, and does whoever reads it know the two differ?
When in your training week was the blood taken, after a hard session or after two easy days?

If the honest answers are no, the general one and nobody knows, then the report hasn't yet answered what you came to it with. That's the whole test. It's the same filter the lactate article applies to every measurement: if a result on either side of the line wouldn't change what you do next, the number is decoration.

Where each ferritin line sits, and who drew it

Ferritin lines in micrograms per litre, from the sources on this page. Each marks something for a population; none is a target for you.
Ferritin line	Who drew it	What it marks
Below 15	WHO, 2020	Iron deficiency in adults who are otherwise well
Below 70	WHO, 2020	Iron deficiency in adults with infection or inflammation
Below 30	Swiss sports medicine consensus, 2015	Low stores in healthy athletes older than fifteen
Above 150 (women), above 200 (men)	WHO, 2020	Risk of iron overload in adults who are otherwise well

Caveats, and where the evidence is thin

A reference page ought to say where it's weak, so here is each limit I know of.

Reference intervals are population statistics. They're built from people presumed healthy, and the laboratory's choice of those people shapes the edges. Nothing on this page says your own range is wrong, only that it answers a narrower question than it seems to.
Ferritin is an acute phase protein. Any single reading mixes storage with inflammation. Thurnham's roughly 30 percent is an average across 32 studies, and the effect in one person on one morning can be larger or smaller.
The bone marrow comparison is observational. Every study in Garcia-Casal's review was observational, and the group with visible marrow iron is small.
The morning-after inference is mine. Peeling's 2009 study had eight people and one session and measured hepcidin in urine; it did not measure ferritin the next morning. Pairing that CRP figure with WHO's rule is my reasoning, and a study drawing ferritin and CRP together on the days after hard training would confirm or break it.
The timing window rests on sixteen people. McCormick's runners all had a ferritin below 50 and ran once at each time of day. Whether the window holds for people with fuller stores, or across a real training week, is open.
Sex and ferritin are tangled in the pooled hepcidin data. Peeling's 2014 low group was mostly women and its high group mostly men.
The null trials are small. Klingshirn had eighteen runners, and Rubeor's own verdict on the twelve trials is that the evidence is equivocal. The cut-off of 20 keeps recurring because the trials that found an effect all enrolled people at or below it, so the positive evidence describes people near empty and says little about someone sitting at 40.
The donor data are American. Cable's donors were screened against the US haemoglobin floor, so the percentages describe that system.

Where Atlas Cove fits

Iron was never my problem. I want that as plain here as it is in the essay, because a page like this one invites the assumption that its author found his answer in a ferritin result somewhere, and I didn't. What turned my bad years around was a change of inputs: I took the alcohol out, kept my sleep to the same hours and trained less hard for some months, or rather I gave up waiting on a number and changed what I was feeding in. Recovery came back first. Performance took longer. What had gone wrong was regulation: several systems working against one another under long stress, which is something no line on a panel reports. A ferritin drawn in a bad week wouldn't have found it. My objection is to how I read the panels, with every value inside a range I'd never questioned and nothing next to it to say what it meant for me.

Our hosts begin every week with a guest's own numbers. Before any of them is read, they settle with you which reference fits your life, and what a high or a low result would actually change in your days. That's the order an Atlas Cove week follows.

Common questions

Can ferritin be normal and still too low for someone who trains?

It can. The bottom of the usual range marks the point where stores are empty, so a value a little above it is inside the lines and still close to empty. Some sports physicians, the Swiss consensus among them, read ferritin against a higher line than the general range, and which one applies to you is worth asking whoever reads your report.

Does hard exercise change ferritin results?

It can make them harder to read. Ferritin rises with inflammation, and in one small study of eight runners a single hard hour left the inflammation marker CRP well above its starting value when blood was drawn again a full day later. I'd want a ferritin drawn the morning after a hard day read next to an inflammation marker, or drawn after a couple of easy days instead.

Does giving blood lower your iron stores?

In a large American donor study, frequent donors were far more likely to have empty stores than first-time donors, and the odds climbed with the number of donations. The check before donating measures haemoglobin, which falls late, so passing it says little about stores.

What should I ask about my ferritin result?

Three things: whether an inflammation marker was measured with it, which range the report uses, and when in your training week the blood was taken. If nobody can answer those, the result hasn't been read yet, only printed.

Sources

Nemeth, E., Tuttle, M. S., Powelson, J., Vaughn, M. B., Donovan, A., Ward, D. M., Ganz, T., & Kaplan, J. (2004). Hepcidin regulates cellular iron efflux by binding to ferroportin and inducing its internalization. Science, 306(5704), 2090-2093. DOI: 10.1126/science.1104742
Peeling, P., Dawson, B., Goodman, C., Landers, G., Wiegerinck, E. T., Swinkels, D. W., & Trinder, D. (2009). Effects of exercise on hepcidin response and iron metabolism during recovery. International Journal of Sport Nutrition and Exercise Metabolism, 19(6), 583-597. DOI: 10.1123/ijsnem.19.6.583
Peeling, P., Sim, M., Badenhorst, C. E., Dawson, B., Govus, A. D., Abbiss, C. R., Swinkels, D. W., & Trinder, D. (2014). Iron status and the acute post-exercise hepcidin response in athletes. PLOS ONE, 9(3), e93002. DOI: 10.1371/journal.pone.0093002
McCormick, R., Moretti, D., McKay, A. K. A., Laarakkers, C. M., Vanswelm, R., Trinder, D., Cox, G. R., Zimmerman, M. B., Sim, M., Goodman, C., Dawson, B., & Peeling, P. (2019). The impact of morning versus afternoon exercise on iron absorption in athletes. Medicine & Science in Sports & Exercise, 51(10), 2147-2155. DOI: 10.1249/MSS.0000000000002026
World Health Organization. (2020). WHO guideline on use of ferritin concentrations to assess iron status in individuals and populations. Geneva: World Health Organization. ISBN 978-92-4-000012-4
Garcia-Casal, M. N., Pasricha, S. R., Martinez, R. X., Lopez-Perez, L., & Peña-Rosas, J. P. (2018). Are current serum and plasma ferritin cut-offs for iron deficiency and overload accurate and reflecting iron status? A systematic review. Archives of Medical Research, 49(6), 405-417. DOI: 10.1016/j.arcmed.2018.12.005
Clénin, G., Cordes, M., Huber, A., Schumacher, Y. O., Noack, P., Scales, J., & Kriemler, S. (2015). Iron deficiency in sports: definition, influence on performance and therapy. Swiss Medical Weekly, 145, w14196. DOI: 10.4414/smw.2015.14196
Sim, M., Garvican-Lewis, L. A., Cox, G. R., Govus, A., McKay, A. K. A., Stellingwerff, T., & Peeling, P. (2019). Iron considerations for the athlete: a narrative review. European Journal of Applied Physiology, 119(7), 1463-1478. DOI: 10.1007/s00421-019-04157-y
Cable, R. G., Glynn, S. A., Kiss, J. E., Mast, A. E., Steele, W. R., Murphy, E. L., Wright, D. J., Sacher, R. A., Gottschall, J. L., Vij, V., & Simon, T. L. (2011). Iron deficiency in blood donors: analysis of enrollment data from the REDS-II Donor Iron Status Evaluation (RISE) study. Transfusion, 51(3), 511-522. DOI: 10.1111/j.1537-2995.2010.02865.x
Thurnham, D. I., McCabe, L. D., Haldar, S., Wieringa, F. T., Northrop-Clewes, C. A., & McCabe, G. P. (2010). Adjusting plasma ferritin concentrations to remove the effects of subclinical inflammation in the assessment of iron deficiency: a meta-analysis. American Journal of Clinical Nutrition, 92(3), 546-555. DOI: 10.3945/ajcn.2010.29284
Klingshirn, L. A., Pate, R. R., Bourque, S. P., Davis, J. M., & Sargent, R. G. (1992). Effect of iron supplementation on endurance capacity in iron-depleted female runners. Medicine & Science in Sports & Exercise, 24(7), 819-824. PMID 1501568
Rubeor, A., Goojha, C., Manning, J., & White, J. (2018). Does iron supplementation improve performance in iron-deficient nonanemic athletes? Sports Health, 10(5), 400-405. DOI: 10.1177/1941738118777488

This is an educational and strategic perspective, not personal medical advice.

The views are the author's own and not statements by Atlas Cove Lda.

See how your week starts with your numbers

Tom Wuerden · Co-Founder

Engineer turned Ironman