Abstract Background This study investigated the indications, procedures and outcomes for adrenal surgery from the UK Registry of Endocrine and Thyroid Surgery database from 2005 to 2017, and compared outcomes between benign and malignant disease. Methods Data on adrenalectomies were extracted from a national surgeon-reported registry. Preoperative diagnosis, surgical technique, length of hospital stay, morbidity and in-hospital mortality were examined. Results Some 3994 adrenalectomies were registered among patients with a median age of 54 (i.q.r. 43–65) years (55·9 per cent female). Surgery was performed for benign disease in 81·5 per cent. Tumour size was significantly greater in malignant disease: 60 (i.q.r. 34–100) versus 40 (24–55) mm (P < 0·001). A minimally invasive approach was employed in 90·2 per cent of operations for benign disease and 48·2 per cent for cancer (P < 0·001). The conversion rate was 3·5-fold higher in malignant disease (17·3 versus 4·7 per cent; P < 0·001). The length of hospital stay was 3 (i.q.r. 2–5) days for benign disease and 5 (3–8) days for malignant disease (P < 0·050). In multivariable analysis, risk factors for morbidity were malignant disease (odds ratio (OR) 1·69, 1·22 to 2·36; P = 0·002), tumour size larger than 60 mm (OR 1·43, 1·04 to 1·98; P = 0·028) and conversion to open surgery (OR 3·48, 2·16 to 5·61; P < 0·001). The in-hospital mortality rate was below 0·5 per cent overall, but significantly higher in the setting of malignant disease (1·2 versus 0·2 per cent; P < 0·001). Malignant disease (OR 4·88, 1·17 to 20·34; P = 0·029) and tumour size (OR 7·47, 1·52 to 39·61; P = 0·014) were independently associated with mortality in multivariable analysis. Conclusion Adrenalectomy is a safe procedure but the higher incidence of open surgery for malignant disease appears to influence postoperative outcomes.