We built and verified a nomogram model for individually predicting patients with T2EL (exposed to age, smoking, hypertension, and number of lumbar arteries) who are more likely to encounter re-intervention. We also determined that an IMA diameter of > 2.77 mm and a < 45.5% proportion of thrombus volume may be risk factors for re-intervention. The nomogram incorporated demographic information, medical history, and aneurysm anatomical characteristics in patients with AAA and showed good discrimination and calibration performance. Thus, it can provide effective assistance for preventing re-intervention in patients with T2EL. Meanwhile, the ROC curve analysis resulted in two novel thresholds which may cause patients with T2EL to consider re-intervention, despite the thresholds exhibiting acceptable differences compared with previously reported values, such as an IMA diameter of 2.5 mm[15] or 3.5 mm[16] and a thrombus volume of < 40.0%[11]. For a patient with AAA diagnosed in the clinic, it is possible to quickly and easily estimate the possibility of re-intervention once T2EL has occurred using the novel nomogram model and thresholds, effectively working as a measure to avoid re-hospitalization and additional financial burdens for these patients, as well as to alleviate the financial deficit of the medical insurance system. Thus, the application of this prediction model to prevent re-intervention of patients with AAAs has both patient and economic benefits.
In a previous study, several risk factors were determined that may increase the incidence of T2EL after EVAR, such as older age, chronic renal failure, chronic obstructive pulmonary disease (COPD), smoking, hypertension, and other anatomical features of the aneurysm[8, 17, 18]. In our predictive model, older age, smoking, and hypertension are also independent risk factors that may increase the incidence of re-intervention in patients with T2EL. In the recent largest prospective study in Japan[8], the researchers examined the medical records of 17,099 patients who underwent EVAR for AAA, and the results showed that age is an independent risk factor for T2EL, which indicates that advanced age has a more important influence on the occurrence of T2EL, in agreement with other previously published studies[7, 19] and our present research. Smoking and hypertension are considered important risk factors for the occurrence and development of AAA, and smoking cessation can effectively reduce the AAA rupture rate by 20%[20]. Meanwhile, smoking and hypertension can also prompt the occurrence of T2EL, as identified by the aforementioned largest prospective study[8], recent meta-analyses [7, 21], and other retrospective studies[16, 22]. Thus, patients with T2EL-related risk factors should be considered when implementing pre-embolism treatment during primary EVAR intervention.
Surgeons often have more interest in the relationship between the feeding arteries of the aneurysm sac, anatomical features of the aneurysm body and neck, thrombus proportion of the aneurysm, and T2EL-related re-intervention. These factors can be treated by endovascular surgery. T2ELs are formed by incompletely occluded branch arteries that continuously provide blood to the gap between the stent graft and the arterial wall; thus, T2EL generally originates from the inferior mesenteric artery, lumbar artery, and accessory renal artery, which are the main branches of an infrarenal AAA. The inferior mesenteric artery has the largest diameter of the main branches from the perspective of previous studies, and an inferior mesenteric artery diameter of > 2.5 mm[15, 23] or > 3.5 mm[16] may improve the incidence of T2EL after EVAR. In addition, an increasing number of patent lumbar arteries may also increase the occurrence of T2EL after EVAR. In a recent study, 5.5 patent lumbar arteries were defined as the threshold that may promote the incidence of T2EL after EVAR[16]. It is widely accepted that the maximum diameter and number of feeding arteries are significant risk factors for T2EL. In our study, we also identified that the diameter and number of feeding arteries are independent risk factors for T2EL-related re-intervention. The T2EL-re-intervention group had a larger diameter of the inferior mesenteric artery, and through ROC curve analysis, the threshold was defined as 2.77 mm. Logically, a larger diameter of the feeding artery yields more difficult occlusion, which can lead to persistent feeding to the gap between the stent graft and the aorta wall, promoting the formation of T2EL. ROC curve analysis showed good sensitivity and specificity. Thus, we have reason to believe that feeding artery diameter is a risk factor for T2EL-related re-intervention, as the diameter of the inferior mesenteric artery was > 2.77 mm. The number of patent lumbar arteries is another risk factor for re-intervention. Recent studies[16, 22, 24] stated that a higher number of patent lumbar arteries, ranging from four to six, may increase the incidence of T2EL. In our study, the results of univariate and multivariate analyses showed a significant statistical difference between the four patent lumbar arteries in the T2EL-related re-intervention group and three patent lumbar arteries in the non-T2EL group, indicating that if patients have a higher number of patent lumbar arteries, they would be more likely to suffer a re-intervention after EVAR. Thus, pre-embolism treatment may be beneficial in these patients. Interestingly, thrombus in the aneurysm sac may be a protective factor against T2EL-related re-intervention. According to a previous study, patients with < 40% thrombus in the aneurysm may prefer re-intervention after EVAR[11]. However, in previous research on the mechanism of AAAs, intraluminal thrombus was defined as a negative role that may aggravate tissue oxidative stress reaction and promote the recruitment of inflammatory factors and lead to the development of AAAs[20]. Thus, it is difficult to define the exact role of intraluminal thrombi in AAAs. In the present study, patients from the T2EL group had a lower proportion of intraluminal thrombus than those from the non-T2EL group, and the ROC curve analysis showed a threshold of 45.5%, which manifested a positive role to prevent re-intervention in the non-T2EL group. The existence of an intraluminal thrombus may assist in occlusion of the branches and increase the tightness between the stent graft and artery wall, thus decreasing the need for T2EL-related re-intervention. In summary, using this prediction model, we proposed that patients with a high prediction probability of re-intervention should undergo pre-embolism treatment in their primary EVAR treatment.