Short term use of HD-DEX has shown less adverse effects (AEs) than PDS maintenance but its success rates reported differently by several studies. The objective of this study was to determine HD-DEX effectiveness in adult ITP with newly diagnosed (group 1) and relapsed after other treatments(Tx) (group 2). 71 (median age; 45, 20~82) with ITP with a platelet counts (PC) 20,000/uL or less than 50,000/uL and clinically significant bleeding were enrolled. DEX a dose of 40 mg/day for 4 consecutive days was tried. A response was defined as an increase in the PC of at least 30,000/uL, PC of more than 50,000/uL by day 10 (D10) after DEX. A maintenance was defined as a PC of more than 50,000/uL 6 ms after Tx. The number of group 1 and 2 were 54, 17, retrospectively. In group 2, previous regimens included;oral PDS and/or IVIG (16)and splenectomy (4). Median PC before Tx were 8,500(1,000~45,000/uL) and 9,000(100~33,000/uL) in group 1 and 2, retrospectively (p=0.80). 27(50%) and 6(35.3%) in group 1, 2 showed good initial response to 1st course of DEX, retrospectively (p=0.40). After 1st course of DEX, 11(20.3%) in group 1 and 1(5.9%) in group 2 could maintain response for 3 ms (p=0.27). By 6 ms, 8 (14.8%) in group 1 and 1 (5.9%) in group 2 kept first response (p=0.68). Among 33 patients bearing initial response, 12 (44.4%)and 2 (33.3%) in group 1, 2 relapsed (p=1.0). Good correlation revealed between PC (D10) and response rate (RR) (p<0.05). Of 21 with PC (D10) above 100,000/uL, 6(28.6%) experienced relapse. On the other hand, nine of 11 patients (81.8%) in whom PC (D10) were less than 100,000/uL relapsed after all. There was no significant difference in relapse-free survival (RFS) after 1st DEX between 2 groups (p=0.9). 13 of 14 relapsed received 2nd DEX and then maintained PDS without interruption. After several weeks of maintenance, PDS tapered according to the PC. 9(75.0%) in group 1 and 2(100%) in group 2 showed response in 2nd DEX. Of these 11 patients, 2 in group 1 relapsed 10 ms after receiving 2nd DEX. After 1st DEX, 12 (16.9%), 5(7%) and 6 (8.5%) temporarily suffered from vague generalized sx., facial edema, and mild liver function abnormality, retrospectively. During the 2nd DEX and consecutive oral PDS, 4(30.7%), 3(23.1%) and 1(7.7%) experienced same AEs, retrospectively. Additionally, of these 13, overt DM requiring hypoglycemic agents appeared in 2(15.4%) and 1(7.7%) experienced duodenal ulcer bleeding. In conclusion, 1 course of HD-DEX revealed good initial response in group 1 but in group 2, the RR was lower than in historical results of oral PDS. However, it has an advantage for the patients to have a good compliance to the Tx because of few AEs and no need to maintain long-term daily medication. Furthermore, although the 2nd DEX and PDS maintenance showed good response, it revealed no benefit compared as a previously established oral PDS regimen because of frequently occurrence of AEs and poor compliance. Therefore, the proper DEX schedule remained to be determined according to the early achieved platelet counts (eg. D10) with 1st course DEX or to the presence of previous Tx.