Background: The risks of local recurrence and treatment-related morbidity need to be balanced after local excision of early rectal cancer. The aim of this meta-analysis was to determine oncological outcomes after local excision of pT1-2 rectal cancer followed by no additional treatment (NAT), completion total mesorectal excision (cTME) or adjuvant (chemo)radiotherapy (aCRT). Methods: A systematic search was conducted in PubMed, Embase and the Cochrane Library. The primary outcome was local recurrence. Statistical analysis included calculation of the weighted average of proportions. Results: Some 73 studies comprising 4674 patients were included in the analysis. Sixty-two evaluated NAT, 13 cTME and 28 aCRT. The local recurrence rate for NAT among low-risk pT1 tumours was 6⋅7 (95 per cent c.i. 4⋅8 to 9⋅3) per cent. There were no local recurrences of low-risk pT1 tumours after cTME or aCRT. The local recurrence rate for high-risk pT1 tumours was 13⋅6 (8⋅0 to 22⋅0) per cent for local excision only, 4⋅1 (1⋅7 to 9⋅4) per cent for cTME and 3⋅9 (2⋅0 to 7⋅5) per cent for aCRT. Local recurrence rates for pT2 tumours were 28⋅9 (22⋅3 to 36⋅4) per cent with NAT, 4 (1 to 13) per cent after cTME and 14⋅7 (11⋅2 to 19⋅0) per cent after aCRT. Conclusion: There is a substantial risk of local recurrence in patients who receive no additional treatment after local excision, especially those with high-risk pT1 and pT2 rectal cancer. The lowest recurrence risk is provided by cTME; aCRT has outcomes comparable to those of cTME for high-risk pT1 tumours, but shows a higher risk for pT2 tumours.