The objective of this study was to evaluate the efficacy of Imatinib on patients with chronic myeloid leukemia (CML) in chronic phase and to analyze its influencing factors. 85 patients received Imatinib mesylate at a dose of 300-600 mg orally per day, and were evaluated for hematologic, cytogenetic, and molecular responses. The results showed that the median follow-up was 21 (range 9 - 78) months. Cumulative complete hematological remission (CHR) rate was 100%, major cytogenetic remission (MCyR) rate was 80%, complete cytogenetic remission (CCyR) rate was 67.1% and complete molecular remission (CMoR) rate was 36.4%. The median time to complete hematological remission (CHR) was 1 (range 1 - 3) month, to complete cytogenetic remission (CCyR) was 6 (range 1 - 24) months. The estimated overall survival rates for patients who received Imatinib for 1, 2, 3 years were (98.7 +/- 1.3)%, (96.5 +/- 2.5)% and (90.1 +/- 6.6)% respectively. The estimated progression-free survival rates at 1, 2, 3 years were (97.6 +/- 1.6)%, (96.1 +/- 2.2)% and (90.0 +/- 1.4)% respectively. The CHR, MCyR and CCyR between low risk, intermediate risk and high risk groups according to the Sokal scoring system and between primarily treated and retreated groups all had no difference. The overall survival of patients who achieved MCyR or CCyR was better than that in patients only achieved hematologic remission (p = 0.026), but there was no significant difference in progression-free survival between them. Univariate analysis for efficacy of Imatinib mesylate revealed that WBC count < 100 x 10(9)/L (p = 0.024), Hb level > or = 130 g/L (p = 0.036), and peripheral basophil count < or = 0.05 (p = 0.024) before therapy were independent favourable factors for achieving MCyR or CCyR. It is concluded that the patients with CML in chronic phase treated with Imatinib can achieve the best hematologic remission and higher cytogenetic remission, it should be considered that the imatinib is a drug of the first-line therapy for untreated and treated patients with CML.