The purpose of our study was to determine if the ethane content of expired air could be a useful index of vitamin E status in children. Eight children with vitamin E deficiency secondary to chronic severe liver disease were studied: six of these children were treated with parenteral vitamin E (2-5 mg/kg/dose every 4-7 d). Measures of vitamin E status pre- and posttherapy were: serum vitamin E, 2 +/- 1 versus 7 +/- 1 micrograms/mL (p less than 0.001); serum vitamin E:total lipids, 0.3 +/- 0.1 versus 1.0 +/- 0.1 mg/g (p less than 0.001); and erythrocyte peroxide hemolysis test, 80 +/- 10 versus 6 +/- 12% (p less than 0.001). Fasting breath ethane in the patients pre- and posttherapy was 78 +/- 10 versus 31 +/- 11 pmol/kg/min (p less than 0.001). Breath ethane correlated negatively with serum vitamin E (p less than 0.042) and serum E:total lipids (p less than 0.004) and positively with the erythrocyte peroxide hemolysis test (p less than 0.003). Values for treated patients did not differ from those for fasted sibling controls (34 +/- 12 pmol/kg/min), postprandial sibling controls (31 +/- 12 pmol/kg/min), and healthy children sampled randomly, in the nonfasted state (21 +/- 14 pmol/kg/min). Breath ethane production in one patient (up to 168 pmol/kg/min) did not normalize after treatment of vitamin E deficiency until her selenium deficiency was corrected as well. We conclude that this noninvasive test can be useful as a screen for vitamin E deficiency in children and for ascertaining response to therapy.(ABSTRACT TRUNCATED AT 250 WORDS)