1 Introduction

Despite not generally being considered as an important test–for example see Blood Chemistry and CBC Analysis which states on page 239 “It is not a helpful test for those who do not have anemia”–RDW has been noted as a useful risk factor in a number of recent papers.

Red blood cell distribution width and mortality risk in a community-based prospective cohort: NHANES III RDW and mortality risk (Perlstein et al. 2009) - In an initial analysis RDW stood out as an important mortality variable. This paper investigates that association and is a good model for doing so. It also provides a reference to Benchmarks for the assessment of novel cardiovascular biomarkers (Morrow and de Lemos 2007) which provides an overview of biomarker assessment.
Also see http://www.biomedcentral.com/1471-2261/13/113
Papers citing the Perlstein RDW paper look like a good place to follow-up.
For example: The role of red blood cell distribution width in mortality and cardiovascular risk among patients with coronary artery diseases: a systematic review and meta-analysis

Red cell distribution width and risk of coronary heart disease events (Zalawadiya et al. 2010) - more RDW, downloaded

Red blood cell distribution width and the risk of death in middle-aged and older adults - more RDW
Note criteria they give for deficiencies on page 3.
“exploratory analyses showed that serum antioxidants, such as carotenoids, selenium, and vitamin E, were strongly associated with RDW, but practically did not change the effect of RDW on mortality”

Red Cell Distribution Width and Depression Among Patients Undergoing Angiography - can’t find actual paper.

The Evaluation of Red Cell Distribution Width in Chronic Hemodialysis Patients (H. Tekce et al. 2014) - looks into associations with high RDW.

Serum antioxidants and inflammation predict red cell distribution width in older women: The Women’s Health and Aging Study I (Semba et al. 2010) - selenium predicts RDW.

Gender and Ethnic Differences in Red Cell Distribution Width and Its Association With Mortality Among Low Risk Healthy United State Adults (Zalawadiya et al. 2012)

Red blood cell distribution width: A simple parameter with multiple clinical applications (Salvagno et al. 2014) - no full text.
“…variety of underlying metabolic abnormalities such as shortening of telomere length, oxidative stress, inflammation, poor nutritional status, dyslipidemia, hypertension, erythrocyte fragmentation and alteration of erythropoietin function.”

Association between red blood cell distribution width and polycystic ovary syndrome (Yilmaz et al. 2014)
Downloaded as 10.3109@07435800.2014.987398.pdf

Red cell distribution width to platelet ratio: New and promising prognostic marker in acute pancreatitis (Çetinkaya 2014) - RDW / Platelets ratio
Also see RDW to Platelet Ratio: A Novel Noninvasive Index for Predicting Hepatic Fibrosis and Cirrhosis in Chronic Hepatitis B (B. Chen et al. 2013) - also see AAR, APRI, FIB4

Red Cell Distribution Width as a Predictor of Prostate Cancer Progression (Albayrak et al. 2014)

Value of Red Cell Distribution Width for Assessing Disease Activity in Crohn’s Disease - no fulltext. cutoff 15.6%
Downloaded 6/1/16 as 10.1097@MAJ.0000000000000334.pdf DOI 10.1097/MAJ.0000000000000334

Red cell distribution width and mean platelet volume in patients with irritable bowel syndrome (Aktas et al. 2014)

Association of erythrocyte deformability with red blood cell distribution width in metabolic diseases and thalassemia trait

Reference intervals of complete blood count constituents are highly correlated to waist circumference: should obese patients have their own “normal values?” (Vuong et al. 2014)

The Red Cell Histogram and The Dimorphic Red Cell Population (Constantino 2011) - beyond RDW. Since RDW calculation involves dividing by MCV would it be worth looking at RDW * MCV?

Clinical evaluation of red cell volume distribution width (RDW)

Red Cell Distribution Width, Revisited (Constantino 2013)

2 Intermountain Risk Score and RDW

One recent use of RDW is in the Intermountain Risk Score.

The IMRS is explained in detail in the paper Exceptional Mortality Prediction by Risk Scores from Common Laboratory Tests (Horne et al. 2009). It is notable that an RDW in the highest quintile (> 14.3%) is assigned the same risk score as the difference between being 70-79 years old and being 40-49! The normal reference range is 11.7 - 15.0% so includes high risk values.

3 Analysis

I have seen little in the RDW literature concerning causes, associations, treatments, etc. Investigate those aspects here.

Here is one paper along those lines: Serum antioxidants and inflammation predict red cell distribution width in older women: the Women’s Health and Aging Study I (Semba et al. 2010)

First load the data.

Loaded saved NHANES III data from NHANES3_20141227.rda

3.1 High RDW Causes and/or Associations

Weatherby indicates that the following conditions are associated with increased RDW levels: iron deficiency, B12/folate deficiency, selenium deficiency
He also lists as related tests: MCV, MCH, HGB, HCT, RBC, serum iron, serum ferritin, serum/urinary methylmalonic acid, homocysteine, TIBC, % transferrin saturation
RDW is included in a number of his anemia screens. See: W.Anemia.B12 and W.Anemia.Iron.Deficiency which use a cutoff of 13.

(Perlstein et al. 2009) hypothesized a role for inflammation and investigated RDW association with CRP.

Look at the following variables: CRP, MCV, MCH, Hemoglobin, Hematocrit, RBC, Homocysteine, Folate.Serum, Folate.RBC, Vitamin.B12, Vitamin.B6 (not in NHANES III), Total.Iron, Ferritin, TIBC, Selenium

I’m having trouble finding NHANES III MMA data, but see Elevated Serum Methylmalonic Acid Concentrations Are Common among Elderly Americans (Morris et al. 2002)
Transferrin.Sat will be available next time I rebuild data.

First just look at the correlations.

We see signs of an inverse association between RDW and MCV, MCH, Hemoglobin, Hematocrit, Total.Iron
but not much else.

Let’s try looking at scatterplots. A number of these curves are U shaped and seem to indicate that RDW is associated with otherwise deranged RBC status (which is unsurprising). I am not seeing much else here. It may help to limit the axes to have more resolution in the areas of interest.

3.2 Possible Derived Biomarkers

Look at the RDW * MCV product. Considering this as a possible biomarker. See (Constantino 2011) “RDW” is what they call RDW-CV. They argue RDW-SD might be a better measure.

x$RDW.MCV <- x$RDW * x$MCV / 100
cor(x$RDW, x$RDW.MCV, use="pairwise.complete.obs")
## [1] 0.7023
scatterplot(x$RDW, x$RDW.MCV, xlab="RDW", ylab="RDW*MCV",
              main="Scatterplot of RDW*MCV/100 vs RDW")

scatterplot(x$RDW, x$RDW.MCV, xlab="RDW", ylab="RDW*MCV",
            xlim=c(11,17), ylim=c(9,17),
              main="Scatterplot of RDW*MCV/100 vs RDW")

There is a fair amount of vertical spread here and I think it is biologically reasonable to think of the width of the distribution as what matters and not the width divided by the mean. Try RDW*MCV/100 out in a COX PH model. After doing this it turns out it is a good metric, but not as good as RDW by itself.

4 To Do

Bibliography

Aktas, Gulali, Aytekin Alcelik, Buket Kin Tekce, Vildan Tekelioglu, Mustafa Sit, and Haluk Savli. 2014. “Red Cell Distribution Width and Mean Platelet Volume in Patients with Irritable Bowel Syndrome.” Gastroenterology Review 3. Termedia Sp. z.o.o.: 160–63. doi:10.5114/pg.2014.43578.

Albayrak, Sebahattin, Kursad Zengin, Serhat Tanik, Hasan Bakirtas, Abdurrahim Imamoglu, and Mesut Gurdal. 2014. “Red Cell Distribution Width as a Predictor of Prostate Cancer Progression.” Asian Pacific Journal of Cancer Prevention 15 (18). Asian Pacific Organization for Cancer Prevention: 7781–4. doi:10.7314/apjcp.2014.15.18.7781.

Chen, Baode, Bo Ye, Jian Zhang, Lixiong Ying, and Yu Chen. 2013. “RDW to Platelet Ratio: A Novel Noninvasive Index for Predicting Hepatic Fibrosis and Cirrhosis in Chronic Hepatitis B.” Edited by James Fung. PLoS ONE 8 (7). Public Library of Science (PLoS): e68780. doi:10.1371/journal.pone.0068780.

Constantino, Benie T. 2011. “The Red Cell Histogram and the Dimorphic Red Cell Population.” Laboratory Medicine 42 (5). Oxford University Press (OUP): 300–308. doi:10.1309/lmf1uy85hekbmiwo.

———. 2013. “Red Cell Distribution Width, Revisited.” Laboratory Medicine 44 (2). Oxford University Press (OUP): e2–e9. doi:10.1309/lmz1gky9lqtvfbl7.

Çetinkaya, Erdinç. 2014. “Red Cell Distribution Width to Platelet Ratio: New and Promising Prognostic Marker in Acute Pancreatitis.” World Journal of Gastroenterology 20 (39). Baishideng Publishing Group Inc.: 14450. doi:10.3748/wjg.v20.i39.14450.

Horne, Benjamin D., Heidi T. May, Joseph B. Muhlestein, Brianna S. Ronnow, Donald L. Lap ’e, Dale G. Renlund, Abdallah G. Kfoury, et al. 2009. “Exceptional Mortality Prediction by Risk Scores from Common Laboratory Tests.” The American Journal of Medicine 122 (6). Elsevier BV: 550–58. doi:10.1016/j.amjmed.2008.10.043.

Morris, Martha Savaria, Paul F. Jacques, Irwin H. Rosenberg, and Jacob Selhub. 2002. “Elevated Serum Methylmalonic Acid Concentrations Are Common Among Elderly Americans.” The Journal of Nutrition 132 (9). Oxford University Press (OUP): 2799–2803. doi:10.1093/jn/132.9.2799.

Morrow, D. A., and J. A. de Lemos. 2007. “Benchmarks for the Assessment of Novel Cardiovascular Biomarkers.” Circulation 115 (8). Ovid Technologies (Wolters Kluwer Health): 949–52. doi:10.1161/circulationaha.106.683110.

Perlstein, Todd S., Jennifer Weuve, Marc A. Pfeffer, and Joshua A. Beckman. 2009. “Red Blood Cell Distribution Width and Mortality Risk in a Community-Based Prospective Cohort.” Archives of Internal Medicine 169 (6). American Medical Association (AMA): 588. doi:10.1001/archinternmed.2009.55.

Salvagno, Gian Luca, Fabian Sanchis-Gomar, Alessandra Picanza, and Giuseppe Lippi. 2014. “Red Blood Cell Distribution Width: A Simple Parameter with Multiple Clinical Applications.” Critical Reviews in Clinical Laboratory Sciences 52 (2). Informa UK Limited: 86–105. doi:10.3109/10408363.2014.992064.

Semba, Richard D., Kushang V. Patel, Luigi Ferrucci, Kai Sun, Cindy N. Roy, Jack M. Guralnik, and Linda P. Fried. 2010. “Serum Antioxidants and Inflammation Predict Red Cell Distribution Width in Older Women: The Womens Health and Aging Study I.” Clinical Nutrition 29 (5). Elsevier BV: 600–604. doi:10.1016/j.clnu.2010.03.001.

Tekce, Hikmet, Buket Kin Tekce, Gulali Aktas, Mehmet Tanrisev, and Mustafa Sit. 2014. “The Evaluation of Red Cell Distribution Width in Chronic Hemodialysis Patients.” International Journal of Nephrology 2014. Hindawi Limited: 1–6. doi:10.1155/2014/754370.

Vuong, Jennifer, Yuelin Qiu, Myanh La, Gwen Clarke, Dorine W. Swinkels, and George Cembrowski. 2014. “Reference Intervals of Complete Blood Count Constituents Are Highly Correlated to Waist Circumference: Should Obese Patients Have Their Own Normal Values?” American Journal of Hematology 89 (7). Wiley-Blackwell: 671–77. doi:10.1002/ajh.23713.

Yilmaz, Calan Mehmet, Sefa Kelekci, and Muzaffer Temur. 2014. “Association Between Red Blood Cell Distribution Width and Polycystic Ovary Syndrome.” Endocrine Research 40 (4). Informa UK Limited: 181–87. doi:10.3109/07435800.2014.987398.

Zalawadiya, Sandip K., Vikas Veeranna, Ashutosh Niraj, Jyotiranjan Pradhan, and Luis Afonso. 2010. “Red Cell Distribution Width and Risk of Coronary Heart Disease Events.” The American Journal of Cardiology 106 (7). Elsevier BV: 988–93. doi:10.1016/j.amjcard.2010.06.006.

Zalawadiya, Sandip K., Vikas Veeranna, Sidakpal S. Panaich, Luis Afonso, and Jalal K. Ghali. 2012. “Gender and Ethnic Differences in Red Cell Distribution Width and Its Association with Mortality Among Low Risk Healthy United State Adults.” The American Journal of Cardiology 109 (11). Elsevier BV: 1664–70. doi:10.1016/j.amjcard.2012.01.396.