Ferritin Levels – Redefining Normality in Women
Medicine must improve because we want to treat the sick better and because we need to keep up with the new diagnoses being created that turn healthy people into patients. Today, Julie Laurence takes us through the coming public health crisis of “low ferritin without anemia.” Not only is ferritin unreliable at diagnosing iron deficiency, but low ferritin without anemia is usually asymptomatic.
Adam Cifu
The American Society of Hematology (ASH) recommends raising the lower limit of normal for ferritin to 30 μg/L for men and women alike. ASH is also proposing a new diagnosis: Iron deficiency without anemia. This diagnosis would be based solely on ferritin values. This follows a trend we see across medicine of turning healthy people into patients.
Increasing the lower limit of normal from 25 to 30 in men won’t change much, but going from 15 to 30 in women and labeling low ferritin as a pathological state would be an important change. According to population prevalence in North America, up to 38% of menstruating women could be defined as having the new condition, iron deficiency without anemia.
ASH has not released screening recommendations, but other hematology associations and experts are already advocating for screening menstruating women, given the high prevalence of this condition. Some labs are already flagging ferritin levels below 30 as abnormal.
We are at a turning point in the overdiagnosis of iron deficiency because ferritin is not a perfect test for it, and iron deficiency without anemia is probably unimportant.
First, the flaws of using ferritin to diagnose iron deficiency. A Cochrane review from 2021 found insufficient evidence to recommend ferritin as a screening tool for iron deficiency in healthy populations. Ferritin is an indirect marker of iron deficiency. To assess true iron deficiency, bone marrow sampling is required. Original studies on ferritin show us how poorly ferritin performs as a test for iron deficiency. In young healthy women specifically, iron depletion with bone marrow staining correlated with a median ferritin level of 9 µg/L – well below the new recommended threshold of 30 µg/L. Furthermore, up to 25% of women were iron replete at a ferritin level of 27 µg/L. Many women would thus be misdiagnosed as iron-deficient using ferritin in this context, especially given the new threshold.
Ferritin levels also fluctuate in young women with menses and with iron-rich food intake. Labeling normal ebbs as a disease state contributes to overmedicalizing normality.
Then there is the issue that treating iron deficiency without anemia has not been shown to be beneficial. Low ferritin levels are poorly correlated with reported symptoms in the absence of anemia – hair loss and restless leg syndrome are the only symptoms that reach statistical significance in a well-designed study. Most notably, reported fatigue does not seem to make the cut. Do we know what we are treating then? How can we measure treatment efficacy in the absence of reliable symptoms?
I’m not minimizing the symptoms or the burden of iron deficiency anemia. Are iron deficiency and anemia prevalent in menstruating women? Yes. Should we treat symptomatic women presenting for care? Absolutely. Should we screen for and treat iron deficiency in the absence of anemia? I’m not so sure.
Ferritins should be ordered and interpreted when there is suspicion of symptomatic iron deficiency anemia (or iron overload). Iron deficiency without anemia should not be a new diagnosis for which healthy women are tested.
Defining ferritin below 30 as being ‘abnormal’ while it is found in 38% of people seems aberrant.
So then, what should normal ferritin levels be? Normal values should be normal by definition, i.e. ‘’occurring naturally’’, ‘’approximating the statistical average or norm’’, or by scientific definition, normals should be the lower and higher limits of a bell curve of what is found in a sample of the general, otherwise healthy, population.
Let’s not make ferritin what it’s not.
Julie Laurence is a registered adult care nurse practitioner in Montreal who specializes in internal medicine at a community hospital outpatient clinic. She is affiliated with the Choosing Wisely campaign and trains future nurse practitioners at both the Université de Montréal and McGill University.
Photo Credit: Filip Mroz
 | A guest post by
|
I understand the concern about medicalization and overdiagnosis. Medicine does plenty of that, sometimes at huge cost and with very little benefit. As a hematologist, I just see this topic differently.
To me, ferritin is the warning zone, like the warning track in baseball. It does not mean the outfielder has already hit the wall. It means the wall is getting closer, and there is still time to slow down. A ferritin of 20 or 25 with a normal hemoglobin still needs context, but I do not think it should be ignored.
There is also a reason 30 keeps coming up as the cutoff. Older single-digit ferritin thresholds were very specific for advanced iron deficiency, but they miss a lot of patients whose iron stores are already depleted. When you look at markers of iron-restricted red blood cell production, the point where the body starts showing changes in RBC production is much closer to 30. Hemoglobin can stay normal for a long time while iron stores are falling, so waiting for anemia means waiting until later in the process.
I see this over and over again. Patients show up after they are already symptomatic and anemic. Some are even getting blood transfusions or expensive IV iron. Some then need urgent GI workups. Then you look back and see a ferritin of 18, 22, or 30 from a year or two earlier that was treated as normal because the hemoglobin had not dropped yet.
The risk-benefit ratio here is very different from many of the things we argue about in medicine. Ferritin is a cheap test. It is not perfect, but when it is low, it is a pretty useful marker of depleted iron stores. The bigger problem with ferritin is usually in the other direction: inflammation can make it look falsely normal or high and hide iron deficiency. A low value is much harder to dismiss.
So when I compare the downside of checking ferritin and acting on a low value in context against the downside of waiting, the balance is not close. On one side is a low-cost blood test and, for many patients, an over-the-counter oral iron supplement. On the other side are ED visits, symptomatic anemia, expensive IV iron, urgent referrals, delayed evaluation for blood loss, and sometimes red blood cell transfusions.
So yes, I agree we should be careful about expanding diagnoses. But in this case, waiting for anemia often creates the very thing we are trying to avoid: more testing, more cost, more invasive care, more anxiety, and more missed chances to intervene earlier. Recognizing low ferritin in context feels like common-sense prevention to me.
This is all the rage among the wellness industry. A thyroid above 1.5, a high uric acid without gout, a low ferritin without anemia, a detectable TPO w normal TSH are all sub-optimal states that can be treated for a low monthly subscription fee. Sorry to see a professional society has jumped on this wellness influencer trend. It is much easier to blame my overall not feeling well on my low ferritin than on the fact that I get 2,000 steps / day and eat pop tarts for every meal.