Anastomotic leakage is a complication of colorectal surgery. C-reactive protein (CRP) is an acute-phase marker that can indicate surgical complications. We determined whether serum CRP levels in patients who had undergone colorectal surgery can be used to exclude the presence of anastomotic leakage and allow safe early discharge. We included 90 patients who underwent colorectal surgery with primary anastomosis. Serum CRP levels were measured retrospectively on postoperative days (PODs) 1 - 7. Patients with anastomotic leakage (n = 11) were compared to those without leakage (n = 79). We statistically analysed data and plotted receiver operating characteristic curves. The incidence of anastomotic leakage was 12.2%. Diagnoses were made on PODs 3 - 24. The overall mortality rate was 3.3% (18.2% in the leakage group, 1.3% in the non-leakage group; P < 0.045). CRP levels were most accurate on POD 4, with a cutoff level of 180 mg/L, showing an area under the curve of 0.821 and a negative predictive value of 97.2%. Lower CRP levels after POD 2 and levels <180 mg/L on POD 4 may indicate the absence of anastomotic leakage and may allow safe discharge of patients who had undergone colorectal surgery with primary anastomosis.