Introduction-aim: Anastomotic leakage (AL) is the most feared complication after colorectal surgery. Its prognosis is related to early recognition and post-operative management and should be based on the concept of "failure to rescue". The aim of the present study was to validate the "DULK" diagnostic score for AL, developed in The Netherlands.
Material and method: From June 2012 to June 2013, 100 consecutive patients were included in a prospective study according to the post-operative care protocol developed by den Dulk et al. The score is based on several laboratory and clinical items recorded daily.
Results: Among the 100 patients, 12 developed a post-operative AL (12%) with a specific mortality rate of 16.6% (2 patients). A DULK-score>3 was good criteria for early diagnosis of AL with a sensitivity of 91.7%, a specificity of 55.7%, a positive predictive value of 22%, a negative predictive value of 98%, and an area under the ROC curve of 0.83. If used routinely, the DULK-score would allow diagnosis of AL 3.5 days earlier than clinical judgment alone. The DULK-score was superior to common diagnostic criteria described in the literature such as the C-reactive protein or procalcitonin, in terms of both sensitivity and specificity.
Conclusion: The DULK-score is, at present, the most reliable instrument for early diagnosis of AL after colorectal surgery and should be integrated into risk management health policies aiming to improve the quality of care according to the "failure to rescue" concept.
Keywords: Anastomotic leakage; Colorectal surgery; Failure to rescue; Post-operative complication; Score.
Copyright © 2014 Elsevier Masson SAS. All rights reserved.