BACKGROUND Glucose management is an important component of peri-operative care. The usefulness of continuous glucose monitoring (CGM) in noncardiac surgery is uncertain. OBJECTIVE To systematically assess the glycaemic profile and clinical outcome of patients equipped with a CGM device during the peri-operative period in noncardiac surgery. DESIGN Systematic review. DATA SOURCES Electronic databases were systematically searched up to July 2024. ELIGIBILITY CRITERIA Any studies performed in the peri-operative setting using a CGM device were included. Closed-loop systems also administering insulin were excluded. Analyses were stratified according to diabetes mellitus status and covered intra-operative and postoperative data. Outcomes included glycaemic profile (normal range 3.9 to 10.0 mmol l−1), complications, adverse events, and device dysfunction. RESULTS Twenty-six studies (1016 patients) were included. Twenty-four studies were not randomised, and six used a control arm for comparison. In bariatric surgery, diabetes mellitus patients had a mean ± SD glucose of 5.6 ± 0.5 mmol l−1, with 15.4 ± 8.6% time below range, 75.3 ± 5.5% in range and 9.6 ± 6.7% above range. During major surgery, diabetes mellitus patients showed a mean glucose of 9.6 ± 1.1 mmol l−1, with 9.5 ± 9.1% of time below range, 56.3 ± 13.5% in range and 30.6 ± 13.9% above range. In comparison, nondiabetes mellitus patients had a mean glucose of 6.4 ± 0.6 mmol l−1, with 6.7 ± 8.4% time below range, 84.6 ± 15.5% in range and 11.2 ± 4.9% above range. Peri-operative complications were reported in only one comparative study and were similar in CGM and control groups. Device-related adverse events were rare and underreported. In 9.21% of cases, the devices experienced dysfunctions such as accidental removal and issues with sensors or readers. CONCLUSION Due to the limited number of controlled studies, the impact of CGM on postoperative glycaemic control and complications compared with point-of-care testing remains unknown. Variability in postoperative glycaemic profiles and a device dysfunction rate of 1 in 10 suggest CGM should be investigated in a targeted surgical group.