Factors pronòstics en la dehiscència anastomòtica de la cirurgia del càncer de còlon. Models de predicció actuals i validació externa independent

INTRODUCTION: Anastomotic leak (AL) is the most dreadful and life-threatening complication in colon cancer (CC) surgery. It is associated to an increase of morbimortality, longer hospital stay, increased healthcare costs and a diminished overall survival rate and a disease free survival rate. The incidence of AL ranges from 3 to 30% depending on the paper, and its mortality ranges from 10 to 15%. Multiple risk factors for AL, related to the patient, the tumor and the treatment, have been described in literature. In addition, different predictive models for AL have been published, but most of them are too complex to be used in daily medical practice. Continuous research on risk factors for AL is crucial and a comfortable and easy to use predicting model for AL is needed, in order to have a better detection of patients at risk and to prevent or minimize risk factors before surgery. OBJECTIVE: The objectives in our study were: (1) to identify risk factors for AL in elective CC surgery at Dr. Josep Trueta de Girona University Hospital (DJTGUH), (2) to perform an external independent validation of current and most important predictive models published in literature and (3) to develop a predicting model for our population. In addition, incidence of AL, clinical impact and survival in our cohort of patients (with and without AL) was studied, and differences between right CC and left CC were sought ; INTRODUCCIÓ: La dehiscència anastomòtica (DA) és la complicació més greu i temuda en la cirurgia del càncer de còlon (CC). S’associa a un augment de la morbimortalitat, l’estada hospitalària i dels costos sanitaris i a una disminució de la supervivència global i lliure de malaltia. La seva incidència varia entre el 3% i el 30% segons les publicacions, i la mortalitat es troba entre el 10% i el 15%. Múltiples factors de risc per a la DA, relacionats amb el pacient, el tumor i el seu tractament, han estat descrits en la literatura. Així mateix, existeixen diferents models de predicció del risc de DA, però la major part d’ells ...