Objectives The aim of this study was to conduct a systematic review to compare the survival rates between short implants (length < 10 mm) versus standard-length implants (≥ 10 mm) inserted in grafted bone. As secondary outcomes, marginal bone loss and survival rates of the implant supported prostheses were also analysed. Materials and methods Randomised controlled trials (RCT) that compared both techniques were searched on three electronic databases till June 2016, a manual search was performed on the bibliography of the collected articles, and the authors were contacted for additional references. The estimates of the interventions were expressed in relative risk (RR), mean implant survival rates and mean differences in marginal bone. Results Eight RCTs were included in this study. From a total of 458 short implants, 15 failed (mean survival rates = 96.7%), While from 488 regular implants, 13 failed (mean survival rates = 97.3%). The technique did not significantly affect: the implant failure rate (P > 0.05), with RR of 1.34 (95% CI 0.67-2.87), the mean differences of marginal bone loss (P = 0.18; MD − 0.04 mm [− 0.10; 0.02] 95% CI), at loading or prosthesis failures rates (RR:0.98; 95% CI 0.40-2.41). The mean differences of marginal bone at 1 year follow-up (post loading) presented significant marginal changes in the short implant group (P = 0.002; MD − 0.10 mm [− 0.16; − 0.03] 95% CI) although a significant high heterogeneity was found between groups. Conclusions This systematic review suggests no difference between both techniques in the treatment of atrophic arches. However, more long-term RCTs are needed to evaluate the predictability at the long run. Clinical relevance The use of short implants might be considered an alternative treatment, since it usually requires fewer surgical phases and tends to be a more affordable option.