What a High RDW on a Blood Test Means

RDW (red cell distribution width) is the coefficient of variation of your red blood cell volumes: the analyzer sizes tens of thousands of individual cells, builds a volume histogram, and reports the spread of that histogram relative to its mean. Most labs report RDW-CV with an upper reference limit between 14.5% and 15.5%, so a value of 15.9% will be flagged high on nearly every report. A high RDW means your red cells are less uniform in size than usual, which happens whenever the bone marrow is producing cells under changing conditions: iron, B12, or folate availability shifting, a recent bleed, hemolysis, a transfusion, or systemic inflammation altering erythropoiesis. A mildly elevated RDW with an otherwise normal complete blood count is common and nonspecific, and the interpretation belongs with a clinician who can see the rest of your panel and your history. What follows is how the number is produced, what it is sensitive to, and how to use it as a longitudinal signal rather than a single flag.

What the analyzer is computing

Modern hematology analyzers such as the Sysmex XN, Beckman Coulter DxH, and Abbott Alinity h-series size red cells one at a time. They do this by impedance as each cell passes through an aperture or by light scatter in a flow cell. The result is a histogram of cell volumes in femtoliters. MCV (mean corpuscular volume) is the mean of that distribution. RDW-CV is the standard deviation divided by the mean, times 100, usually computed over a truncated window of the histogram to exclude platelet clumps at the low end and doublets at the high end. RDW-SD is a different statistic: the width of the histogram in femtoliters at the 20% height level, reported directly in fL with a typical range near 39 to 46 fL.

The distinction matters when you pull results programmatically. RDW-CV is normalized by MCV, so a macrocytic sample (high MCV) can have a wide absolute histogram and a normal-looking RDW-CV. RDW-SD does not have that dependency. In HL7 or FHIR data, RDW-CV usually carries LOINC 788-0 and RDW-SD carries LOINC 21000-5. The units also differ, % versus fL, so a parser that keys on the display name “RDW” alone will eventually concatenate two incompatible series. Key on the LOINC code and the UCUM unit, and reject any observation where they disagree.

Why 15.9% is flagged, and why the cutoff is not universal

Reference intervals for RDW-CV are set per analyzer and per population, and the truncation windows and calibration differ between manufacturers. A value of 15.9% sits above the upper limit on essentially all common platforms, so it will carry an H flag. A value of 14.6% might be flagged at one lab and normal at another. RDW-CV is not harmonized across platforms the way hemoglobin roughly is, which means a change from 13.4% to 14.8% between two draws at different labs is not interpretable as a biological change. Record the performing lab and analyzer with every result, and only compute deltas within a single method.

Within a single method, RDW is a stable individual trait. Analytical imprecision is typically around 1 to 2% CV, and within-person biological variation over weeks is small, so your own running baseline is a far tighter reference than the population interval. A person who sits at 12.4% and moves to 14.3% has changed substantially even though both values are inside the reference range. This is the main argument for keeping your own CBC time series instead of reading each report in isolation.

Reading RDW with MCV, hemoglobin, and reticulocytes

RDW is close to uninformative alone and quite informative in combination. The standard framing pairs it with MCV. A low MCV with a high RDW describes an anisocytic microcytic picture. It is most often seen with iron deficiency, where a population of small, newly made cells mixes with older normal ones. A low MCV with a normal RDW describes a uniformly small red cell population, the pattern typical of thalassemia trait. There the Mentzer index (MCV divided by RBC count, below 13) is often used as a first discriminator. A high MCV with a high RDW is the pattern seen with B12 or folate deficiency, brisk reticulocytosis, or liver disease. Large young cells or macrocytes coexist with normal ones.

None of these mappings is diagnostic. The tests below are what separate them, and ordering and interpreting that workup is a clinician’s job.

• iron studies (ferritin, transferrin saturation, soluble transferrin receptor)
• B12 and methylmalonic acid
• reticulocyte count and immature reticulocyte fraction
• a peripheral smear

If you are looking at a new or rising RDW with anemia, that is the conversation to have rather than a supplement to buy.

Why RDW correlates with outcomes across unrelated conditions

The reason RDW appears in so many prognostic models has less to do with red cells than with what disturbs them. In a large cohort of unselected outpatients, RDW correlated significantly with high-sensitivity CRP and erythrocyte sedimentation rate. That supports reading it partly as a downstream marker of chronic inflammation.¹ Inflammatory cytokines blunt erythropoietin responsiveness, impair iron availability, and shorten red cell survival. All of these widen the volume distribution over the weeks it takes the circulating pool to turn over.

That mechanism explains the breadth of the association literature. In a general population followed for cardiovascular events, higher RDW predicted cardiovascular morbidity and all-cause mortality independent of standard covariates.² In acute decompensated heart failure, RDW predicted short- and long-term outcomes more effectively than hemoglobin did.³ It carries prognostic weight in chronic heart failure over two years of follow-up.⁴ The same holds in pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension.⁵ It also holds in patients presenting to the emergency department with acute infection, where it tracked alongside established prognostic biomarkers.⁶ Reviews of the cardiovascular literature treat it as a nonspecific integrator of inflammation, oxidative stress, and nutritional status rather than a marker of any single pathway.⁷

The practical consequence is that RDW has good population-level prognostic value and poor individual-level specificity. It tells you something is perturbing erythropoiesis. It does not tell you what.

Handling RDW in your own data

We would treat RDW as a slow-moving trend variable and ignore single-point flags near the boundary. Concretely, store each CBC as a row with the following fields.

• LOINC code
• value
• unit
• collection timestamp
• performing lab
• analyzer model

Plot RDW-CV, MCV, hemoglobin, and reticulocyte count on a shared time axis, because RDW changes are only interpretable next to the parameters that drive them. Compute a per-person mean and SD from at least four or five draws on the same method before you call anything a deviation. Use a reference change value rather than the population interval. With analytical CV near 1.5% and small within-person biological variation, a change of roughly 10% or more from your own baseline is worth noticing, while a 0.2 percentage point wobble is noise.

Two preanalytical failure modes will generate spurious values. Cold agglutinins cause red cell clumping at room temperature, which the analyzer sizes as very large particles and which distorts both MCV and RDW. Warming the sample to 37 °C and re-running resolves it. Prolonged storage of EDTA whole blood causes cells to swell, raising MCV and widening the histogram, so samples analyzed many hours after draw drift upward. A recent transfusion mixes two donor populations and mechanically raises RDW for weeks with no bearing on your own marrow. Check for these before interpreting a new high value.

Questions people also ask

Is an RDW of 15.9 considered high? Yes, on essentially every common analyzer, since upper reference limits fall between about 14.5% and 15.5%. Whether it is meaningful depends on your hemoglobin, MCV, reticulocyte count, and your own prior values.

Is it better to have high or low RDW? Lower values within the reference range are associated with better outcomes in population studies, and there is no condition in which a low RDW itself is a problem. A very low RDW simply means uniformly sized red cells.

Can dehydration increase RDW? Dehydration concentrates plasma and raises hemoglobin, hematocrit, and red cell count. RDW-CV is a ratio of the volume distribution’s spread to its mean, so it is largely unaffected by plasma volume. A short-term RDW jump is more likely from a sample handling problem or a real change in erythropoiesis.

What can I do to bring my RDW down? RDW is an output and makes no sense as a target. It falls when whatever is widening the distribution resolves, which requires identifying the cause through iron, B12, folate, reticulocyte, and inflammatory testing with a clinician. Treating the number directly is not a coherent goal.

What level of RDW is concerning? There is no single threshold. Studies typically split cohorts around 14.5% to 15.5% and find graded risk above that, but the useful signal for an individual is a sustained rise from your own baseline, especially alongside a falling hemoglobin or a changing MCV.

Footnotes

1. Giuseppe Lippi, Giovanni Targher, Martina Montagnana, et al. Relation Between Red Blood Cell Distribution Width and Inflammatory Biomarkers in a Large Cohort of Unselected Outpatients. Archives of Pathology & Laboratory Medicine, 2009. https://doi.org/10.5858/133.4.628 ↩

2. Dahlia Weitzman, Raanan Raz, Arie Steinvil, et al. Red blood cell distribution width and the risk of cardiovascular morbidity and all-cause mortality. Thrombosis and Haemostasis, 2014. https://doi.org/10.1160/th13-07-0567 ↩

3. YUXIANG DAI, HAKUOH KONISHI, ATSUTOSHI TAKAGI, et al. Red cell distribution width predicts short- and long-term outcomes of acute congestive heart failure more effectively than hemoglobin. Experimental and Therapeutic Medicine, 2014. https://doi.org/10.3892/etm.2014.1755 ↩

4. Łukasz Wołowiec, Daniel Rogowicz, Joanna Banach, et al. Prognostic significance of red cell distribution width and other red cell parameters in patients with chronic heart failure during two years of follow-up. Kardiologia Polska, 2016. https://doi.org/10.5603/kp.a2016.0004 ↩

5. Anna Smukowska-Gorynia, Iga Tomaszewska, Katarzyna Malaczynska-Rajpold, et al. Red Blood Cells Distribution Width as a Potential Prognostic Biomarker in Patients With Pulmonary Arterial Hypertension and Chronic Thromboembolic Pulmonary Hypertension. Heart, Lung and Circulation, 2018. https://doi.org/10.1016/j.hlc.2017.08.007 ↩

6. Giuseppe Lippi, Lorenzo Teti, Mariella Dipalo, et al. Relationship between red blood cell distribution width and prognostic biomarkers in patients admitted to the emergency department with acute infections. European Journal of Internal Medicine, 2013. https://doi.org/10.1016/j.ejim.2012.09.005 ↩

7. Mesay Arkew, Kabtamu Gemechu, Kassahun Haile, et al. Red Blood Cell Distribution Width as Novel Biomarker in Cardiovascular Diseases: A Literature Review. Journal of Blood Medicine, 2022. https://doi.org/10.2147/jbm.s367660 ↩