The purpose of this randomized clinical study was to compare the effect of 2 surgical methods in the treatment of gastric dilatation-volvulus (GDV) in dogs. One group of dogs (group A) was treated with and one group (group B) without fixation of the stomach. Group A consisted of 21 cases (including 2 dropouts) and group B of 10 cases. The dogs in group A received decompression, anatomical repositioning of the stomach and a circumcostal gastropexy and the dogs in group B (the control group) received the same treatment without gastropexy. Supportive treatment was the same for both groups. The randomization of the dogs in groups A and B was successful with only small differences between the 2 groups in the breed, age, sex and initial decompression methods. At the end of the study (censoring time), the median survival times were significantly different between group A and group B, respectively 549 and 107 days. There were no recurrences in group A while in group B 3 dogs (50%) experienced a recurrence within 6 months. The overall death rates within the first year were 32% in group A and 80% in group B. The death rates caused by GDV and GDV related causes only, after one year of follow-up, were 19% and 71% for groups A and B, respectively. This study shows that treatment that included circumcostal gastropexy significantly reduced the recurrence of GDV and prolonged the postoperative survival time compared with treatment that did not include fixation of the stomach.