Physicians are moving away from routine axillary lymph node dissection (ALND) in clinically node-negative breast cancer. We conducted a systemic review on the safety of this policy. Pubmed and Cochrane library were searched for. Sixty-eight studies were included: studies of clinically node-negative patients in the pre-sentinel node (SN) era; observational studies of SN-negative patients, without ALND; comparative studies of SN-negative patients, with a non-ALND and an ALND group; SN-positive studies, of patients without ALND. Primary endpoint was the pooled axillary recurrence rate (ARR) of each category; secondary endpoint was overall survival (OS) rate. In pre-SN studies, with larger tumors and less systemic therapy, ARR without ALND after 5-10 years follow-up was 12-18%, with 5% reduced OS. In the observational SN-negative studies, with median follow-up of 36 months, the pooled ARR was 0.6% (95% CI 0.6-0.8). In the comparative SN-negative studies, pooled ARR was 0.4% (95% CI 0.2-0.6) without ALND versus 0.3% (95% CI 0.1-0.6) with ALND at 31 and 47 months, respectively, and no survival disadvantage. In SN-positive studies, ARR was up to 1.7% (95% CI 1.0-2.7) at 30 months. For patients with an H&E positive SN the ARR without ALND was 5% after 23 months, which may imply rates as high as 13 and 18% after 5 and 8 years. In conclusion, this systematic review confirms the safety of omitting ALND in SN-negative patients. There is a potential role for avoiding ALND in selected SN-positive patients, but eligibility criteria and the role of systemic therapy need further to be elucidated.