The addition of bevacizumab to standard chemotherapy has improved progression-free survival (PFS) and overall survival (OS) in patients with metastatic colorectal cancer (mCRC) in both first- and second line treatment, but the role of maintenance bevacizumab remains controversial. The association of various clinical factor and survival was examined in this retrospective cohort analysis. The clinical data from 220 previously untreated patients with mCRC, not progressive at the end of standard chemotherapy plus bevacizumab, were collected and analyzed. Patients were classified into two subgroups: those given with maintenance bevacizumab: "maintenance bevacizumab cohort (n = 118; MB)", and those discontinuing bevacizumab as a result of physician's or patient's decision: "no maintenance bevacizumab cohort (n = 102; noMB)". The baseline factors were well balanced between the study subgroups. Median PFS and OS for the general population was 10 months (range 7-15) and 22.5 months (range 18-26), respectively. Median PFS was 13 and 8 months in the BM and noBM cohorts, respectively (p < 0.0001). In the multivariate analysis, maintenance therapy resulted independently associated with improved PFS (HR 1.73; p < 0.001), but only objective response (OR) after first-line chemotherapy was associated with improved OS. Maintenance chemotherapy cannot be considered a standard of care after induction chemotherapy for mCRC, because the optimal balance between efficacy and safety of maintenance therapy remains a significant challenge. The results of our retrospective study suggest that maintenance therapy with bevacizumab is a safe and valuable option, particularly in those patients achieving an objective response after first-line chemotherapy.