Université Paris Cité, France & Service de Radiologie, Hôpital Beaujon, APHP Nord, Clichy, France.
Intestinal Stroke Centre, Service de Gastroenterologie, MICI et Insuffisance Intestinale, Hôpital Beaujon, APHP Nord, Clichy, France.
Eur J Vasc Endovasc Surg. 2023 Jun;65(6):802-808. doi: 10.1016/j.ejvs.2023.01.041. Epub 2023 Jan 31.
The aim of this study was to propose computed tomography angiography (CTA) based anatomical segmentation of the superior mesenteric artery (SMA), in order to standardise the reporting of occlusive lesions in acute mesenteric ischaemia (AMI).
A retrospective CTA evaluation of patients with occlusive AMI admitted between 2016 and 2021. After the screening of 468 patients, 95 were included. The SMA was segmented into proximal (S1, ostium to the inferior pancreaticoduodenal artery), middle (S2, from the inferior pancreaticoduodenal to the ileocolic artery), and distal (S3, downstream the ileocolic artery) sections. The jejunal arteries were labelled J1 to J6, and the middle, right, and ileocolic arteries C1, C2, and C3. Two radiologists independently applied the proposed segmentation to a cohort of patients with occlusive AMI to describe occlusive lesions. Intra- and inter-rater agreement was assessed with kappa statistics.
Occlusions involved one segment in 50 (53%) patients (S1, n = 27 [28%]; S2, n = 12 [13%]; S3, n = 11 [12%]); two segments in 37 (39%) patients (S2/S3, n = 31 [33%]; S1/S2, n = 3 [3%]; S1/S3, n = 3 [3%]); and all three segments in eight patients (S1/S2/S3, 8%). The median number of jejunal arteries was four (interquartile range 3, 4.5). C1 and C2 were present in 93 (98%) and 23 patients (24%), respectively. Almost perfect intra-rater agreement was obtained for S1 (91% agreement, κ = 0.82, 95% confidence interval [CI] 0.72 - 0.92); substantial agreement was obtained for S2 (90% agreement, κ = 0.80, 95% CI 0.68 - 0.92) and S3 (86% agreement, κ = 0.72, 95% CI 0.58 - 0.86). Almost perfect inter-rater agreement (with the second junior reading) was obtained for S1 (97% agreement, κ = 0.95, 95% CI 0.89 - 1.0), S2 (91% agreement, κ = 0.82, 95% CI 0.72 - 0.92), and S3 (agreement 96%, κ = 0.91, 95% CI 0.83 - 0.99).
A standardised CTA based anatomical segmental description of SMA occlusive lesions in AMI is proposed; it provided substantial to almost perfect intra- and inter-rater agreement for most anatomical segments.
本研究旨在提出基于计算机断层血管造影术（CTA）的肠系膜上动脉（SMA）解剖分段方法，以便标准化急性肠系膜缺血（AMI）闭塞性病变的报告。
回顾性 CTA 评估 2016 年至 2021 年期间因闭塞性 AMI 入院的患者。在筛选了 468 例患者后，纳入 95 例。将 SMA 分为近端（S1，从起始部到胰十二指肠下动脉）、中段（S2，从胰十二指肠下动脉到回结肠动脉）和远端（S3，回结肠动脉下游）三个节段。空肠动脉标记为 J1 至 J6，中间动脉、右动脉和回结肠动脉标记为 C1、C2 和 C3。两位放射科医生独立地将所提出的分段应用于一组闭塞性 AMI 患者，以描述闭塞性病变。使用 Kappa 统计评估内部和外部评估者之间的一致性。
50 例（53%）患者存在一段闭塞（S1，n=27[28%]；S2，n=12[13%]；S3，n=11[12%]）；37 例（39%）患者存在两段闭塞（S2/S3，n=31[33%]；S1/S2，n=3[3%]；S1/S3，n=3[3%]）；8 例（8%）患者存在三段闭塞（S1/S2/S3，n=8[8%]）。空肠动脉的中位数为 4 根（四分位间距 3，4.5）。C1 和 C2 分别存在于 93（98%）和 23 例患者（24%）中。S1 的内部评估者之间的一致性几乎为完美（91%的一致性，κ=0.82，95%置信区间[CI]为 0.72-0.92）；S2（90%的一致性，κ=0.80，95%CI 为 0.68-0.92）和 S3（86%的一致性，κ=0.72，95%CI 为 0.58-0.86）的一致性为高度一致。与第二位初级阅片医生的外部评估者之间的一致性几乎为完美（97%的一致性，κ=0.95，95%CI 为 0.89-1.0），S1（97%的一致性，κ=0.95，95%CI 为 0.89-1.0）、S2（91%的一致性，κ=0.82，95%CI 为 0.72-0.92）和 S3（96%的一致性，κ=0.91，95%CI 为 0.83-0.99）的一致性为高度一致。
提出了一种基于 CTA 的 SMA 闭塞性病变的解剖分段标准描述方法；对于大多数解剖节段，内部和外部评估者之间的一致性为高度一致到几乎完美。
