Details

Autor(en) / Beteiligte

• Horne, Benjamin D
• Muhlestein, Joseph B
• Bennett, Sterling T
• Jensen, Kurt R
• Marshall, Diane
• McCubrey, Raymond
• Bunch, T J
• Bair, Tami L
• LE, Viet T
• Lappé, Donald L
• Anderson, Jeffrey L
• Knowlton, Kirk U

Titel

Abstract 10605: Skewness but Not Kurtosis of Red Cell Corpuscular Volume Nullifies the Predictive Ability of Red Cell Distribution Width for Predicting All-Cause Mortality

Ist Teil von

• Circulation (New York, N.Y.), 2019-11, Vol.140 (Suppl_1 Suppl 1), p.A10605-A10605

Ort / Verlag

by the American College of Cardiology Foundation and the American Heart Association, Inc

Erscheinungsjahr

2019

Quelle

EZB Electronic Journals Library

Beschreibungen/Notizen

• IntroductionThe red cell distribution width (RDW) is the coefficient of variation of red blood cell (RBC) corpuscular volume (CV) and a powerful predictor of mortality. The RDW size distribution (RDW-sd) and combined extreme erythrocyte macrocytic (EEMa) and microcytic (EEMi) percentages are mortality predictors independent of RDW. This study evaluated whether the skewness of CV (SCV) or kurtosis of CV (KCV) is superior to RDW, RDW-sd, or EEMa/EEMi for predicting mortality.MethodsIntermountain Healthcare patients aged ≥18 years (n=38,279 females, n=30,625 males) had the histogram of RBC sizes, RDW, and RDW-sd collected at the first complete blood count (CBC) in March 2015 through August 2016. The histogram gave counts of RBCs sized 0-250 fL in 5 fL-wide bins from which SCV, KCV, EEMa, EEMi, and other RBC CV measures were calculated. Patients were excluded for blood transfusion within the prior 60 days or pregnancy. Sex-specific Cox regression evaluated associations with mortality adjusting for RDW, RDW-sd, EEMa, EEMi, age, CBC parameters, encounter type (outpatient, emergency, inpatient), and 33 comorbidities.ResultsFemales averaged 49.1±19.7 years of age and males 51.0±18.8 years. Quartiles of SCV, KCV, other CV metrics (i.e., SD, SE, median absolute deviation, average absolute deviation, and median CV), mean CV, RDW, RDW-sd, and the combination of EEMa/EEMi predicted mortality for females and males. In multivariable analysis for females, SCV (quartile 1 vs. 4HR=1.82, 95% CI=1.19, 2.78, p=0.006), EEMa+EEMi (quartiles 4/4 vs. 1/1HR=2.39, 95% CI=1.32, 4.33, p=0.004), and mean CV retained significance but not RDW, RDW-sd, KCV, or others. For males, SCV (quartile 1 vs. 4HR=1.64, 95% CI=1.13, 2.37, p=0.009) and EEMa+EEMi (quartiles 4/4 vs. 1/1HR=2.39, 95% CI=1.48, 3.84, p<0.001) were the multivariable mortality predictors.ConclusionsSCV predicted mortality and eliminated RDW and RDW-sd from multivariable Cox models, becoming the only known factor to nullify the predictive ability of RDW. EEMa/EEMi also predicted mortality. Lower SCV, an indicator of shifted RBC size histogram toward more small RBCs (i.e., a longer or broader left tail or imbalanced histogram), should be investigated for biological meaning and clinical value.