Objective To investigate the state of early marginal bone resorption of different implants in posterior-tooth areas and to explore the related influencing factors. Methods The clinical data were retrospectively analyzed in 96 patients undergoing posterior implant restoration with bone tissue level implants (bone tissue level implant group) and 104 patients with soft tissue level implants (soft tissue level implant group) from Feb. 2013 to Feb. 2016. All patients underwent X-ray examination immediately and at 12 weeks after implantation. With the changes of implants marginal alveolar bone heights as the evaluation basis of early marginal bone resorption, the early mesial and distal marginal bone losses were compared between the two groups. The differences in periodontal health indexes (plaque index[PLI], bleeding index[SBI] and probing depth[PD]) and marginal fitness of implants at 12 weeks after implantation were recorded and analyzed. Pain score (numerical rating scale[NRS]) within 4 weeks after implantation were compared between the two groups. The 200 patients were divided into excellent group (bone resorption<1.5 mm) and poor group (bone absorption ≥ 1.5 mm) according to the state of early marginal bone loss. The general data, including age, gender, location of missed teeth, local health of implant, implant diameter, implant length, implanting methods, and smoking, were compared between the two groups. The independent risk factors of early marginal bone resorption of implants were analyzed by logistic regression analysis. Results There were no significant differences in early mesial or distal marginal bone losses, periodontal health indexes (PLI, SBI, or PD) or grade A rate of marginal fitness of implants between the bone tissue level implant group and soft tissue level implant group at 12 weeks after implantation (all P>0.05); there was no significant difference in NRS score between the two groups within 4 weeks after implantation (P>0.05). The differences in age, gender, location of missed teeth, implant diameter, implant length and other general data were not statistically significant between the excellent group (n=148) and the poor group (n=52, all P>0.05). Logistic regression analysis showed that non-embedded implanting, poor local health environment of implant and smoking were independent risk factors of early implant marginal bone resorption (all P<0.05). Conclusion Bone tissue level implant and soft tissue level implant have similar early marginal bone resorption and high value in posterior-teeth implant restoration. Non-embedded implanting, poor local health environment of implant and smoking are closely related to early marginal bone resorption and need to be taken into account.