What Is RDW-SD in Blood Work and Why Does It Matter?

RDW-SD stands for red blood cell distribution width–standard deviation, and it measures how much your red blood cells vary in size. Specifically, it captures the width of your red cell volume distribution curve at a fixed point, expressed in femtoliters (fL), typically with a normal range around 39 to 46 fL depending on the laboratory. Unlike its more commonly reported cousin, RDW-CV, this measurement is not influenced by average cell size, which makes it a more reliable gauge of true size variation in several clinical situations. That distinction turns out to matter far beyond anemia workups.

How RDW-SD Differs From RDW-CV

Most complete blood count (CBC) reports include a value labeled simply “RDW,” which is almost always RDW-CV, the coefficient of variation. RDW-CV is calculated as a percentage: it divides the standard deviation of red cell volumes by the mean cell volume (MCV). That ratio makes it convenient but also introduces a blind spot. Because MCV sits in the denominator, anything that shifts the average size of your red cells also changes the RDW-CV reading, even if the actual spread of cell sizes has not changed at all.

RDW-SD sidesteps that problem. It is a direct measurement of the width of the red cell volume distribution curve taken at a level 20% above the baseline of the histogram. It is reported in femtoliters, not as a percentage, and it does not depend on MCV at all. In conditions where MCV runs high, such as chronic liver disease, RDW-CV can underestimate how varied the cells really are, while RDW-SD reflects that heterogeneity more accurately.¹ There is a second technical advantage: RDW-CV captures cell sizes within one standard deviation of the mean, which means it misses the extremes. RDW-SD sweeps a broader range of cell volumes, picking up unusually large or small cells that RDW-CV would overlook.

In practice, both numbers appear on many modern automated analyzers, but clinicians have traditionally relied on RDW-CV because it has been around longer and has more published reference data. Research over the last decade is shifting that habit. A study of patients with chronic hepatitis B found that RDW-SD had a notably stronger correlation with liver fibrosis stage than RDW-CV did, with a correlation coefficient more than double that of RDW-CV.² When the researchers compared the two measurements’ ability to diagnose significant fibrosis, RDW-SD produced a larger area under the curve (about 0.71 versus 0.60 for RDW-CV), confirming its superiority in that clinical context.

Reference Ranges and What Shifts Them

A typical reference range for RDW-SD falls roughly between 39 and 46 fL, though exact cutoffs vary by analyzer and laboratory. For RDW-CV, the usual range is about 11.5% to 14.5%. Reference intervals can differ between men and women. A population study in Bangladesh found that RDW-CV intervals were somewhat wider in females (about 11.7–15.4%) than in males (11.7–14.7%), reflecting greater baseline variability in women’s red cell populations.³

Pregnancy is another context where RDW shifts significantly. A systematic review reported that median RDW-CV rises as pregnancy progresses, and the upper reference limit can climb above 19% in the third trimester, compared to around 16–17% in the first trimester. The median RDW value in pregnant women was significantly higher than in non-pregnant women (about 13.5% versus 12.3%).⁴ Despite RDW being in routine clinical use for decades, pregnancy-specific reference ranges remain poorly standardized, which means a slightly elevated RDW in a pregnant patient may be perfectly normal but could be flagged as abnormal by a laboratory using standard cutoffs.

Sorting Out Types of Anemia

One of the oldest clinical uses of RDW is distinguishing between different causes of anemia when red cells are small. Iron deficiency anemia and thalassemia trait both produce small red cells, but they do so for different reasons, and they call for very different management. In iron deficiency, the body makes red cells of wildly varying sizes because iron supply fluctuates, driving RDW up sharply. In thalassemia trait, red cells are uniformly small due to a genetic hemoglobin defect, so RDW tends to stay closer to normal.

Multiple studies have built on that principle. One found that an RDW-CV cutoff above 21% was useful for differentiating between iron deficiency and non-transfusion-dependent thalassemia, with a sensitivity around 84% and specificity around 71% in a validation set.⁵ A pediatric study evaluating a dozen different red cell indices ranked the RDW Index as the second most reliable tool for separating iron deficiency from beta-thalassemia trait in children, with a correct diagnosis rate of 82%.⁶

Machine learning is now pushing this diagnostic task further. A recent study trained five algorithms on CBC parameters, including both RDW-SD and RDW-CV, to distinguish thalassemia trait from iron deficiency. The best-performing model achieved an area under the curve of 0.96 in internal validation and 0.99 in an external cohort. Interpretability analysis showed that RDW-SD was one of the two most influential predictors, alongside mean corpuscular hemoglobin concentration.⁷ The fact that automated systems lean on RDW-SD so heavily underscores its discriminative value compared to the traditional RDW-CV alone.

Why RDW Predicts Mortality

Over the past fifteen years, a surprising research finding has emerged: RDW is one of the strongest routine lab predictors of death from almost any cause, not just blood-related conditions. A large general-population study found that people in the highest quarter of RDW values had about a 34% higher risk of dying from any cause compared to those in the lowest quarter. The association held across specific causes, including cardiovascular disease, cancer, and respiratory disease.⁸

In critically ill patients the relationship is even starker. An analysis of over 26,000 ICU patients showed that each one-unit increase in RDW was associated with roughly a 10–13% increase in the risk of dying at 30 days, 90 days, and one year.⁹ A separate multicenter study measuring RDW during the first week of an ICU stay found that patients with high RDW had roughly double the one-year mortality risk after adjusting for other factors.¹⁰ Intriguingly, the predictive power was actually stronger in patients whose illness was milder at admission, suggesting that a high RDW in someone who does not look very sick may be a particularly important warning sign.

Nobody believes that oddly sized red blood cells themselves cause heart attacks or organ failure. The working explanation is that RDW captures several damaging processes simultaneously: chronic inflammation, oxidative stress, nutritional deficiency, and impaired kidney function all independently drive red cell size variation, and they also independently damage the cardiovascular system and other organs. A high RDW, in other words, is a single cheap number that reflects a web of harmful biology happening in the background.¹¹

Acute Coronary Syndrome and Heart Disease

Beyond general mortality, elevated RDW has specific links to cardiovascular events. A prospective study of patients admitted with acute coronary syndrome found that those with high RDW-CV had an in-hospital mortality rate of about 8%, compared to 1% in the low RDW-CV group.¹² The connection between red cell variation and atrial fibrillation is also increasingly recognized. Inflammation and oxidative stress can simultaneously impair how red cells are produced and disrupt the electrical properties of heart tissue, creating a scenario where a high RDW and heart rhythm problems emerge together. Some evidence goes further, suggesting that anisocytic (unevenly sized) red blood cells may themselves contribute to plaque buildup and adverse cardiac remodeling, making them an active player rather than a passive marker.¹¹

Liver Fibrosis and Kidney Disease

Liver disease is one of the clearest areas where RDW-SD outperforms RDW-CV. In patients with chronic hepatitis B, both measurements rose as fibrosis worsened, but RDW-SD tracked the progression more tightly. Its diagnostic accuracy for identifying significant fibrosis (stages F2 through F4) was meaningfully higher than that of RDW-CV.² The reason ties back to the MCV problem: chronic liver disease tends to push red cells larger, which inflates the denominator in the RDW-CV formula and masks the true heterogeneity. RDW-SD, being independent of MCV, gives a cleaner picture.

In kidney disease, the story unfolds along similar lines. RDW increases in a dose-response fashion as kidney function declines: patients with the most advanced chronic kidney disease (CKD) stages have the highest RDW values, with a clear linear trend across CKD stages 1 through 5.¹³ This makes biological sense: failing kidneys produce less erythropoietin, the hormone that drives red cell production, leading to inefficient and uneven cell maturation. A study found that patients with RDW above 14.5% had more than five times the odds of rapid kidney function deterioration compared to those with lower RDW, even after accounting for other risk factors.¹⁴ Combining RDW with albumin levels into a composite ratio further improved prediction of progression to end-stage kidney disease requiring dialysis.¹⁵

Inflammation and Autoimmune Conditions

Chronic inflammation is probably the single most common driver of elevated RDW outside of nutritional deficiencies. Inflammatory molecules interfere with red cell production in the bone marrow, shortening cell lifespan and producing cells of inconsistent size. This makes RDW a surprisingly useful proxy for inflammatory burden, particularly in diseases where inflammation fluctuates and is hard to track cheaply.

In rheumatoid arthritis, RDW rises alongside standard inflammatory markers like C-reactive protein and erythrocyte sedimentation rate, and it also correlates with disease activity scores. Studies have found that RDW levels track with pro-inflammatory molecules like TNF-alpha and IL-6, while being inversely related to the anti-inflammatory molecule IL-10. Patients with active bone erosion tended to have higher RDW values.¹⁶

A similar pattern appears in systemic lupus erythematosus. Patients with active lupus had a mean RDW of about 16.6%, compared to roughly 13.2% in those with inactive disease and 12.7% in healthy controls. The difference was large enough that the researchers proposed RDW as a practical and inexpensive surrogate marker for disease activity.¹⁷ In settings where more expensive inflammatory markers are not readily available, a simple CBC that already reports RDW can give clinicians a useful signal about whether a flare may be underway.

Cancer Prognosis

Elevated RDW before treatment has emerged as a prognostic marker in multiple cancer types. A meta-analysis pooling data from over 1,000 patients with blood cancers found that high pretreatment RDW predicted poorer overall survival (roughly 2.4 times the risk), poorer progression-free survival, and poorer event-free survival. The pattern held across subtypes including multiple myeloma and diffuse large B-cell lymphoma.¹⁸

In solid tumors, the picture is consistent. A study of pancreatic cancer patients found that those with both jaundice and elevated RDW (RDW-CV above 15.4% or RDW-SD above 46 fL) had a median survival of just 5.6 months, compared to 15.3 months in patients with neither. After adjusting for tumor stage, chemotherapy, and tumor markers, the combination of jaundice and high RDW carried roughly six times the mortality risk.¹⁹ The pancreatic cancer study is notable for using RDW-SD explicitly alongside RDW-CV to define the high-RDW phenotype, reflecting the growing clinical acceptance of RDW-SD as a legitimate and sometimes preferable measurement.

Sepsis and the ICU

In sepsis, where rapid risk stratification can change management, RDW has performed impressively. A study comparing RDW to lactate, the traditional go-to marker for sepsis severity, found that RDW had better discriminative ability for predicting 30-day hospital mortality. Patients with elevated RDW had significantly higher rates of ICU admission (about 49% versus 32%) and septic shock (about 39% versus 24%) compared to those with normal RDW.²⁰ Combining RDW with hemoglobin into a ratio further improved prediction. One analysis identified a hemoglobin-to-RDW ratio below 0.78 as an inflection point: below that threshold, each unit increase in the ratio corresponded to a substantial reduction in 28-day mortality.²¹

The appeal of RDW in critical care is partly practical. It is already on every CBC, costs nothing extra, and requires no additional blood draw. A lactate level also needs a blood sample but is a dedicated test that may not be ordered immediately on every patient. RDW, by contrast, is sitting in the results whether anyone asks for it or not.

Diabetes and Metabolic Complications

In people with diabetes, RDW rises in step with the severity of complications. One study found a clear gradient: healthy controls averaged an RDW-CV of about 12.7%, diabetic patients without complications averaged about 13.9%, and those with complications averaged about 15.3%. Patients with macrovascular damage, such as peripheral vascular disease or coronary artery disease, had the highest mean RDW at around 15.8%. RDW also correlated positively with hemoglobin A1c and urine albumin-to-creatinine ratio, two standard markers of glycemic control and early kidney damage, and correlated negatively with estimated kidney filtration rate.²² After adjusting for A1c, diabetes duration, and kidney function, an RDW of 14.5% or above remained an independent predictor of diabetic complications, with more than three times the odds.

This is clinically relevant because it suggests RDW adds information beyond what standard diabetes monitoring already captures. A patient with a well-controlled A1c but a creeping RDW might still be accumulating vascular damage detectable through the red cell population before it shows up on other tests.

How Exercise Temporarily Raises RDW

If you get a blood test shortly after a long run or other intense endurance exercise, your RDW may look abnormal even though nothing is wrong. A study tracking runners after a moderate endurance event found that RDW peaked about 20 hours after the finish, with a median increase of about 2.2%. The odds of having a RDW above the normal upper limit were nearly four times higher at that time point compared to baseline.²³ The effect is transient and reflects mechanical stress on red cells during prolonged exercise, not any underlying disease. For anyone scheduling a routine CBC, being aware that a hard workout the day before can bump RDW up enough to trigger an abnormal flag is worth knowing to avoid unnecessary follow-up testing.

What to Do When Your RDW-SD Is Flagged

A high RDW-SD on a lab report is not a diagnosis. It is a signal that red cell production or survival is uneven, and the job becomes figuring out why. Your doctor will typically look at the rest of the CBC first: if hemoglobin is low, the question shifts to what kind of anemia is present. If MCV is low with a high RDW, iron deficiency is a strong candidate. If MCV is high with a high RDW, vitamin B12 or folate deficiency enters the picture. If all the standard blood counts look normal but RDW is elevated, the clinical significance depends heavily on context: a person with known inflammatory disease or chronic kidney disease may simply be showing a marker of their underlying condition, while in an otherwise healthy person, a persistently elevated RDW is worth investigating further.

The research is clear that elevated RDW carries prognostic weight across a remarkably wide range of conditions, from common chronic diseases to acute critical illness. But it remains a nonspecific marker. It cannot tell you what is wrong any more than a fever can; it tells you that something is generating enough biological stress to disrupt red cell homeostasis. Its value lies in its ubiquity, low cost, and the growing evidence that it adds predictive information beyond what traditional markers provide. For clinicians, the message is to stop ignoring the RDW line on the CBC printout. For patients, a high RDW-SD is a reasonable prompt to ask your doctor whether it fits a pattern worth investigating.