Advertisement

Left ventricular hypertrophy and diastolic dysfunction in children with sickle cell disease are related to asleep and waking oxygen desaturation

Mark C. Johnson, Fenella J. Kirkham, Susan Redline, Carol L. Rosen, Yan Yan, Irene Roberts, Jeanine Gruenwald, Jan Marek and Michael R. DeBaun

Data supplements

  • Supplemental materials for: Johnson et al

    Files in this Data Supplement:

    • Table S1. Study subjects (PDF, 46.2 KB) -
      Characteristics of study subjects with echocardiogram versus the SAC subjects without echocardiogram.

    • Table S2. Univariable comparisons (PDF, 114 KB) -
      Univariable Spearman Rank Correlations with Bonferroni multiple comparison adjusted p-values for echocardiographic variables versus sleep and saturation variables for 44 study subjects. Correlations with adjusted p-value<0.05 shown in italics

Article Figures & Data

Tables

  • Table 1

    Echocardiographic measures

    Category/variableMean (SD)Expected normal valuesP
    LV size and mass
        LV diastolic dimension Z-score2.5 (1.3)0 (−2 to 2)< .001
        LV mass index, g/m2.744.9 (10.7)26-39*< .001
        Relative wall thickness0.27 (0.04)< 0.41.7
    LV systolic function
        Shortening fraction0.37 (0.04)≥ 0.28.7
    LV tissue Doppler
        E′15.7 (3.4)16.5 (5.3).1
        A′4.8 (1.2)6.4 (1.9)< .001
        S′8.0 (2.1)9.3 (3.4)< .001
        E/E′7.2 (2.2)6.1 (2.4).002
        MPI index0.45 (0.09)0.35 (0.09)< .001
    Septal tissue Doppler
        E′11.7 (2.0)12.6 (3.4).004
        A′5.6 (1.9)6.1 (1.6).1
        S′7.6 (1.1)7.6 (1.9).8
        MPI index0.47 (0.08)0.35 (0.09)< .001
    • Mean (SD) values for echocardiographic variables in 44 study subjects with P values based on comparison with expected normal values.37,40,41 Tissue Doppler values E′, A′, and S′ are given as centimeters per second.

    • MPI indicates myocardial performance index (a measure of combined systolic and diastolic function).

    • * LV mass index normal values vary by age and sex; 27% of the reference population was African American.37

  • Table 2

    Multivariable models with oxygen saturation asleep

    Dependent variable/independent covariatesParameter estimateStandardized coefficientP
    LV mass index, g/m2.7 (R2 = 0.66, P < .001)
        Oxygen saturation asleep, percentage−2.1−0.60< .001
        Age, y−0.73−0.26.02
        Systolic blood pressure percentile0.100.24.03
        Arousal index0.480.18.09
    LV diastolic dimension Z-score (R2 = 0.74, P < .001)
        Oxygen saturation asleep, percentage−0.18−0.43< .001
        Sleep maintenance efficiency, percentage−0.06−0.48< .001
        Systolic blood pressure percentile0.020.42.001
        Heart rate−0.04−0.38.006
        Oxygen desaturation index0.030.32.007
        Body mass index percentile−0.01−0.25.02
        Arginine-ornithine ratio0.220.22.05
    LV E/E′ (R2 = 0.4, P < .001)
        Oxygen saturation asleep, percentage−0.46−0.63< .001
        Oxygen desaturation index−0.04−0.31.03
    • Multivariable models of LV mass index, LV diastolic dimension Z-score, and LV E/E′ with oxygen saturation asleep, other sleep variables, and confounding independent variables for 44 study subjects. Sleep variables as potential independent variables: arousal index, sleep maintenance efficiency, OAHI at 3% desaturation, average oxygen saturation asleep, and oxygen desaturation index. Potential confounding variables evaluated: hemoglobin, LDH, arginine-ornithine ratio, age, heart rate, systolic and diastolic blood pressure percentiles, and body mass index percentile.

  • Table 3

    Multivariable models with waking oxygen saturation

    Dependent variable/independent and confounding variablesParameter estimateStandardized coefficientP
    LV mass index, g/m2.7 (R2 = 0.62, P < .001)
        Waking oxygen saturation, percentage−1.6−0.57< .001
        Age, y−0.78−0.28.02
        Arousal index0.650.24.04
        Systolic blood pressure percentile0.10.23.05
    LV diastolic dimension Z-score (R2 = 0.72, P < .001)
        Sleep maintenance efficiency, percentage−0.07−0.58< .001
        Waking oxygen saturation, percentage−0.13−0.42< .001
        Heart rate−0.04−0.4.006
        Arousal index0.090.28.02
        Systolic blood pressure percentile0.020.31.02
        Hemoglobin, g/dL−0.3−0.28.03
    LV E/E′ (R2 = 0.31, P = .002)
        Waking oxygen saturation, percentage−0.31−0.56< .001
        Oxygen desaturation index−0.05−0.32.04
    • Multivariable models of LV diastolic dimension Z-score and LV mass index with waking oxygen saturation, sleep variables, and confounding independent variables for 44 study subjects. Sleep variables as potential independent variables: arousal index, sleep maintenance efficiency, OAHI at 3% desaturation, and oxygen desaturation index. Potential confounding variables evaluated: hemoglobin, LDH, arginine-ornithine ratio, age, heart rate, systolic and diastolic blood pressure percentiles, and body mass index percentile.