Clinical value of dual low-dose computed tomography angiography incorporating artificial intelligence iterative reconstruction for assessing arteriovenous fistulas/grafts among hemodialysis patients.
Authors
Affiliations (3)
Affiliations (3)
- Department of Radiology, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, China.
- United Imaging Healthcare, Shanghai, China.
- Medical Imaging International Scientific and Technological Cooperation Base of Zhejiang Province, Hangzhou, China.
Abstract
Upper extremity computed tomography angiography (CTA) plays an important role in evaluating arteriovenous fistula or arteriovenous graft (AVF/AVG) dysfunction in hemodialysis patients, but conventional protocols are associated with substantial radiation exposure and contrast load, raising concerns in this vulnerable population requiring repeated imaging. The purpose of this study was to investigate the clinical value of dual low-dose upper extremity CTA incorporating artificial intelligence iterative reconstruction (AIIR) for AVF/AVG assessment in hemodialysis patients. One hundred and two patients (mean age, 60.33±13.60 years, 52 men) with suspected AVF/AVG dysfunction were prospectively enrolled and randomly assigned to undergo either routine-dose CTA (Group A, 100 kVp, 1.0 mL/kg contrast, hybrid iterative reconstruction; n=51) or low-dose CTA (Group B, 80 kVp, 0.6 mL/kg contrast, AIIR; n=51). Qualitative and quantitative results were analyzed and compared. Among these patients, only 44 patients (176 segments) underwent digital subtraction angiography (DSA) serving as the reference standard. The diagnostic performance for detecting significant stenosis (>50%) was evaluated using receiver operating characteristic analysis, intraclass correlation coefficient (ICC), and generalized estimating equations (GEEs). Demographic characteristics were comparable between two groups (all P>0.05). Group B reduced radiation and contrast doses by 52.2% and 40.5%, respectively, compared to Group A. Qualitative scores showed no significant difference between the two groups (all P>0.05). Quantitatively, Group B exhibited approximately 165% and 162% improvement in signal-to-noise ratio and contrast-to-noise ratio over Group A, respectively (all P<0.001). Per-segment and per-patient analyses demonstrated high diagnostic accuracy (98% and 95%, respectively), with excellent CTA-DSA agreement for stenosis grading (ICC >0.90 across all evaluated segments). GEE analysis showed no significant difference between groups in detecting significant stenosis (odd ratio =0.99; 95% confidence interval: 0.86-1.14; P=0.899). Dual low-dose CTA with AIIR significantly reduces radiation and contrast doses, while providing superior image quality and preserving diagnostic confidence for evaluating AVF/AVG stenosis.