Late anastomotic leakage happens early enough: a biochemical argument
Martin Rutegård, Josefin Segelman, Peter Matthiessen, Marie-Louise Lydrup, Jennifer ParkAbstract
Introduction
A substantial share of anastomotic leakage after anterior resection for rectal cancer is diagnosed late in the postoperative period. It is often assumed that these leaks in reality occur earlier. We aimed to investigate this assumption using the systemic inflammatory response in terms of C-reactive protein (CRP) elevation.
Methods
This retrospective cohort study included patients who underwent anterior resection for rectal cancer at 11 centres between 2014 and 2018 with one-year follow-up concerning anastomotic leakage. Exposure was diagnosis of an early leak (≤14 days) compared to late leak (>14 days), while outcome was the maximum CRP level within the first 14 postoperative days. Linear regression modelling with adjustment for confounding was performed.
Results
Some 1118 patients were included. Of these, 109 (9.8%) patients had a leakage within 14 days, while 122 (10.9%) were diagnosed later. Median maximum CRP was 108.5 g/L (IQR: 61–169) in patients without leakage, compared to 270 g/L (IQR: 200–370) and 167.5 g/L (IQR: 120–237) in patients with an early and late leakage, respectively (P = 0.0001). Compared with patients without leakage, the adjusted CRP increase was 132 g/L (95% c.i.: 113–152) for early leaks and 42 g/L (95% c.i.: 24–60) for late leaks.
Discussion
Maximum CRP within the first two postoperative weeks was markedly higher in patients with an anastomotic leakage occurring later than 14 days, in comparison to patients without any leakage. These findings can be interpreted as corroborating the hypothesis that at least some of these later diagnosed leaks in fact were earlier leaks, not yet detected.