A new score to predict post operative complications after endovascular treatment of thoraco abdominal aortic aneurysms.
Acta Chir Belg. 2014 Jul-Aug;114(4):250-5
Authors: Guillou M, Bianchini A, Hertault A, D’Elia P, Pottrain N, Sobocinski J, Maurel B, Azzaoui R, Tyrrell M, Haulon S
BACKGROUND: The endovascular treatment of thoraco abdominal aortic aneurysms (TAAA) is a minimally invasive solution. However, patient selection remains a major problem. We have analysed our experience to identify the risk factors for post-operative morbidity and mortality and to construct a scoring system to identify those patients likely to benefit from this treatment.
METHODS: We have analysed a consecutive cohort of patients treated electively for TAAA using endovascular techniques between 2006 and October 2012. All data were collected prospectively. The risk factors associated with spinal cord ischemia (SCI), the need for post-operative dialysis and 30 day mortality were determined using multivariate statistical techniques and a logistic regression model including all variables that were significant on univariate analysis (p < 0.05). A predictive score was calculated using a Received Operating Characteristic (ROC) curve, defining best specificity and sensibility.
RESULTS: We analysed the data from 123 patients (median age 70 years). The 30 day mortality rate was 8% (10 patients). The SCI rate was 6% (7 patients). One patient (1%) required long-term dialysis after the aortic procedure. The cumulative early mortality, SCI and permanent dialysis rate was 14% (17 patients). In multivariate analysis, adverse outcome was associated with advanced age (OR = 1.110 ; p = 0.022), and Crawford type I or II or III (OR = 9.292 ; p = 0.002) as compared with Crawford type IV. Pre-operative beta blocker (BB) treatment was a protective factor (OR = 0.099 ; p = 0.005). A predictive score was then constructed : Score = -10.060 + 0.104x(A) +2.229x(B) -2.315x(C) (A = patient age ; B = 1 if TAAA Crawford type 1, 2 or 3, 0 if TAAA type 4 ; C = 1 if on-going BB treatment (30 days pre-surgery minimum), 0 if none). Its sensitivity and specificity were 88% and 89% respectively.
CONCLUSIONS: We propose a simple predictive scoring system. This tool is useful in predicting the most feared complications after endovascular TAAA repair and has potential use in the identification and counselling of vulnerable patients being considered for surgery. More data are needed to refine the prediction of individual operative risks.
PMID: 26021420 [PubMed – in process]