Evaluation of infrapopliteal occlusive disease with contrast-enhanced MR angiography in diabetic patients / 中华放射学杂志

ABSTRACT

Objective To evaluate the diagnostic accuracy of contrast-enhanced MR angiography (CE MRA) in infrapopliteal occlusive diseases of diabetic patients.Methods A total of 105 patients with known diabetes and peripheral vascular occlusive disease who underwent both CE MRA and DSA examnations were included in this study.They had no obvious stenosis or with stenosis of less than 75％ in iliac,femoral,and popliteal arteries.Every infra-popliteal artery was anatomically divided into 9 vascular segments as tibiofibular trunk artery,proximal and distal anterior tibial artery,proximal and distal posterior tibial artery,proximal and distal peroneal artery,plantar and dorsalis pedis artery.There were 945 segments.The arterial stenosis was accessed with CE MRA and DSA respectively.The segments were scored in 5 categories as＜30％,≥30％ and＜50％,≥50％ and＜75％,≥75％ and＜100％,and 100％ according to stenostic degrees.The Kappa test was used to compare the diagnostic consistency of CE MRA and DSA.Taken DSA as a gold standard for reference,ROC curve was drawn to calculate the diagnostic accuracy of CE MRA in diagnosis of lower limb arterial disease.Results The ratio of statistically valid segments for both CE MRA and DSA were 97.7％ (923/945) in 945 vessel segments of 105 patients,and 390 of 923 arterial segments were both indicated as occlusion by CE MRA and DSA.The diagnostic consistency for the segments was listed in the decreasing order as followsproximal peroneal artery,proximal posterior tibial artery,distal posterior tibial artery,proximal anterior tibial artery,distal anterior tibial artery,tibiofibular trunk artery,dorsalis pedis artery,plantar artery,distal peroneal artery,and corresponding Kappa values were 0.88,0.86,0.84,0.84,0.81,0.77,0.75,0.75 and 0.73,P＜ 0.05.The AUC(area under ROC curve) of CE MRA was 0.893 with 95％ confidence interval of 0.882 to 0.904.Conclusion CE MRA is an accurate imaging modality in the diagnosis of infrapopliteal occlusive disease s for diabetes.