C-reactive protein can predict anastomotic leak in colorectal surgery: a systematic review and meta-analysis.

Abstract

Anastomotic leakage (AL) is one of the most significant complications after colorectal surgery, affecting length of stay, patient morbidity, mortality, and long-term oncological outcome. Serum C-reactive protein (CRP) level rises in infective and inflammatory states. Elevated CRP has been shown to be associated with anastomotic leak. Perform a meta-analysis of current CRP data in AL after colorectal surgery. MEDLINE, EMBASE, CINAHL, CENTRAL databases STUDY SELECTION: Comparative studies studying serum CRP levels in adult patients with and without AL after colorectal surgery. Elective and emergency open, laparoscopic or robotic colorectal excisions for cancer and benign pathology. Mean serum CRP measurements between post-operative days (POD) 1 through 7 in patients with and without AL. Perform ROC analysis to determine cut-off CRP values to indicate AL. Twenty-three studies with 6647 patients (482 AL). Pooled mean time to diagnosis of AL was 7.70days. AL associated with higher CRP on POD1 (mean difference (MD) 15.19, 95% CI 5.88-24.50, p = 0.001), POD2 (MD 51.98, 05% CI 37.36-66.60, p < 0.00001), POD3 (MD 96.92, 95% CI 67.96-125.89, p < 0.00001), POD4 (MD 93.15, 95% CI 69.47-116.84, p< 0.00001), POD5 (MD 112.10, 95% CI 89.74-134.45, p< 0.00001), POD6 (MD 98.38, 95% CI 80.29-116.46, p < 0.00001), and POD7 (MD 106.41, 95% CI 75.48-137.35, p < 0.00001) compared with no AL. ROC analysis identified a cut-off CRP of 148mg/l on POD3 with sensitivity and specificity of 95%. On POD4 through POD7, cut-off levels were 123mg/l, 115mg/l, 105mg/l, and 96mg/l, respectively, with sensitivity and specificity of 100%. Study heterogeneity, some characteristics unreported, no RCT CONCLUSIONS: AL is associated with higher CRP levels on each post-operative day compared to no AL after colorectal surgery. The cut-off CRP values can be used to predict AL to expedite investigation and treatment.