One of the most serious presentations of preeclampsia is the HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome. Treatment has generally been limited to supportive measures and management of complications, but recent clinical trials—which were not double-blind or placebo-controlled—suggest that dexamethasone may stabilize the disorder and even improve it in the antepartum period, maximizing postpartum recovery. The present randomized, double-blind, placebo-controlled trial enrolled 132 women, 60 of them pregnant and 72 puerperal, who met criteria for complete HELLP syndrome. They were randomly assigned to a treatment or placebo group and in the former case received 10 mg dexamethasone intravenously at 12-hour intervals until delivery and three further doses after delivery. The puerperal women received three doses of dexamethasone. All women received, in addition, magnesium sulfate intravenously and, when needed, an antihypertensive agent, initially nifedipine and subsequently clonidine or amlodipine if necessary to lower the diastolic blood pressure. Hospital time was less for women given dexamethasone, although not to a significant degree. The treatment and placebo groups were similar in the time needed to reach a platelet count exceeding 100,000/mm3. There also was no difference in the proportion of women reaching a lactate dehydrogenase level less than 600 U/L before discharge. Treatment did not hasten recovery in patients with an aspartate aminotransferase level less than 70 U/L before discharge. Urine output was comparable in the treatment and control groups. Three of four maternal deaths were in steroid-treated women. No substantial differences were evident when pregnant and puerperal women were analyzed separately. These findings do not support using dexamethasone to treat HELLP syndrome in preeclamptic women who are in late pregnancy or the early postpartum period.