For patients with dento-maxillofacial deformities who receive orthodontic-orthognathic combined treatment, the conventional treatment approach is preoperative orthodontic-orthognathic surgery-postoperative orthodontics. However, with the development of techniques used in orthodontic and orthognathic treatment, the surgery-first approach (SFA), namely, orthodontic surgery-postoperative orthodontics, has been widely used currently and displays several advantages, such as improving the treatment efficiency and providing patients with more satisfaction. This review provides a brief discussion and review of SFA concerning its development, indications, advantages and disadvantages, outcomes and stability, and the application and research progress of SFA in orthodontic-orthognathic combined treatment for patients with maxillofacial deformity. The literature review results showed that compared with the conventional treatment approach, SFA has relatively strict indications, which usually include patients with skeletal class Ⅱ/Ⅲ malocclusion, skeletal open bite, and bimaxillary protrusion or patients with facial asymmetry but who require little preoperative orthodontic treatment or removal of the compensation of the dental arch, specifically as follows: ①well-aligned to mildly crowded anterior teeth, ②flat to mild curve of Spee, ③normal to mildly proclined/retroclined incisor inclination, ④acceptable arch coordination, ⑤extensive occlusal contact between the upper and lower dentition requiring at least 3 occlusal contacts. Any occlusion that may affect the outcome of surgery or final result of the overall treatment, as well as any disease that may jeopardize the healing process after surgery, is regarded as a contraindication. Furthermore, SFA has potential disadvantages, such as a possible higher incidence of complications, including unstable occlusion and malunion of bones, which still require further research to be confirmed. Most researchers believe that no significant difference occurs between the outcome and stability of the two approaches. However, currently, we still need a sufficient sample size of prospective studies to provide accurate evidence.