Correlation of Baseline Plasma and Inguinal Connective Tissue Metalloproteinases and Their Inhibitors With Late High-Pressure Endoleak After Endovascular Aneurysm Repair: Long-term Results.

To investigate whether plasma and connective tissue matrix metalloproteinases (MMP) and their inhibitors (TIMP) may predict late high-pressure endoleak after endovascular aneurysm repair (EVAR). Samples of inguinal fascia and blood were collected in 72 consecutive patients (mean age 73.1 years; 68 men) undergoing primary EVAR with the Endurant stent-graft. Baseline plasma levels of MMP-2, MMP-9, TIMP-1, and TIMP-2 and baseline MMP-2 and MMP-9 activity estimated using gelatin zymography (GZ) were compared between patients who developed late endoleak in follow-up and those who did not. Subgroup analyses were performed between patients with (n=18) and without inguinal hernias and between patients with moderate-diameter (50-59 mm; n=45) or large-diameter (≥60 mm; n=27) abdominal aortic aneurysms (AAA) at primary EVAR. The mean follow-up period was 63.1 months (range 7.5-91.5), during which time 13 (18.1%) patients developed type I (6 Ia and 5 Ib) or 2 type III endoleaks. Only GZ-analyzed proMMP-9 concentrations were higher in the endoleak group than in patients without endoleak (mean difference 8.44, 95% CI -19.653 to -1.087, p=0.03). The patients with primary inguinal hernia at presentation had significantly higher tissue TIMP-2 values (0.8±0.7 vs 0.5±0.4, p=0.018) but lower plasma total (pro- + active) MMP-9 values (11.9±7.8 vs 16.2±7.4, p=0.042) than patients without hernias at the time of EVAR. Patients with AAAs ≥60 mm had significantly higher mean tissue homogenate levels of total (pro- + active) MMP-9 (p=0.025) and total (pro- + active) MMP-2 (p=0.049) as well as higher proMMP-9 (p=0.018) and total (pro- + active) MMP-9 (p=0.021) levels based on GZ compared to patients with moderate-diameter AAAs. Regression analysis revealed a significant association between total (pro- + active) MMP-9 plasma samples and the presence of hernia (OR 0.899, 95% CI 0.817 to 0.989, p=0.029) and between GZ-analyzed proMMP-9 and late endoleak (OR 1.055, 95% CI 1.007 to 1.106, p=0.025). GZ-analyzed proMMP-9 and active MMP-9 were strong predictors of late endoleak in patients with hernia (p=0.012 and p=0.044, respectively) and in patients with AAAs ≥60 mm (p=0.018 and p=0.041 respectively). Inguinal fascial tissue proMMP-9 significantly predicted late endoleak. ProMMP-9 and active MMP-9 biomarkers are significantly associated with late endoleak in hernia patients and in patients with AAAs ≥60 mm. Considering the clinical association between hernia and AAA and the fact that the AAA wall connective tissue environment remains exposed to systemic circulation after EVAR, inguinal fascia extracellular matrix dysregulation and altered MMP activity may reflect similar changes in AAA biology, leading to complications such as endoleak.