Application of color-coded quantitative digital subtraction angiography in predicting the outcomes of immediate type
Min Zhou1, Zijie Su1, Zhenyu Shi1
1Department of Vascular Surgery, Zhongshan Hospital, Institute of Vascular Surgery, Fudan University, Shanghai, China.
Objective:
Immediate type I and type III endoleaks after endovascular aneurysm repair (EVAR) could be persistent or temporary. Reintervention is necessary for persistent ones. Color-coded quantitative digital subtraction angiography (CQDSA) could provide a quantitative evaluation of the endoleak hemodynamics. We aimed to use CQDSA to quantify immediate type I and type III endoleaks after EVAR and to find a practical way to predict their outcomes.
Methods:
Between January 2012 and December 2014, 485 consecutive patients with abdominal aortic aneurysms underwent EVAR at our institution. Thirty-five patients (31 men, four women) with slight immediate type I and type III endoleaks after EVAR were recruited in the prospective observational nested case-control study. After at least 6 months of follow-up, these patients were divided into two groups based on endoleak-related adverse events. Their final intraprocedure DSA images were collected and converted into a single polychromatic image for CQDSA measurements. The parameter time to peak (TTP) of the selected regions of interest in the endoleak area and a reference area at the same latitude within the stent graft were derived from the time-intensity curve. A receiver operating characteristic curve was generated to test the ability of TTP to predict endoleak-related adverse events and to identify the optimal cutoff value.
Results:
Finally, two groups were identified: 12 patients with endoleak-related adverse events and 23 patients without endoleak-related adverse events. Median follow-up time for all patients was 24.0 months. Age, gender, and comorbidity were similar in these two groups. TTP was significantly lower in patients with endoleak-related adverse events (P = .002). The risk of endoleak-related adverse events was significantly higher in patients with mixed-type endoleak than in those with simple-type endoleak (P = .003). According to the receiver operating characteristic curves, TTP ≤5 seconds reached the maximal sum of sensitivity and specificity (sensitivity, 91.67%; specificity, 69.57%). Logistic regression analysis confirmed that TTP ≤5 seconds (P = .016) and mixed-type endoleak (P = .044) were associated with higher risk of endoleak-related adverse events.
Conclusions:
CQDSA could help predict the outcomes of immediate type I or type III endoleaks after EVAR. TTP ≤5 seconds and mixed-type endoleak were two potential predictors of endoleak-related adverse events. This approach may offer an objective assessment of such immediate endoleaks and reference for immediate reintervention or conservative therapy.
