Objective

The Goutallier classification system is the most commonly used method for grading intramuscular fatty infiltration in rotator cuff tears. This grading system presents low inter-observer reliability and an inability to provide quantitative and repeatable outcomes for intramuscular fat. We determined the correlation and reliability of two methods, the Lava Flex and Ideal IQ MRI techniques, in quantifying volumetric intramuscular fat, while also comparing to the Goutallier method.

Materials and methods

The supraspinatus muscles of seventeen cadaveric shoulders were scanned using the Lava Flex and Ideal IQ MRI imaging protocols. Histological analysis was performed on the same muscles. Agreement, reliability, and correlation analyses were performed to compare all outcomes.

Results

The Lava Flex protocol took an average of ~ 4 min, while the Ideal IQ required about ~ 11 min to complete. Bland–Altman analysis showed good agreement between the Lava Flex and Ideal IQ [LOA (− 0.10 and 0.05)], and ICC analyses showed excellent reliability (ICC (1,1) 0.948; ICC (2,1) 0.947). There was a 91% correlation between the Lava Flex and Ideal IQ MR protocols. Weighted Kappa analysis between histology and the Goutallier classification showed fair-to-moderate agreement.

Discussion

The Lava Flex technique, taking about 30% of the acquisition time, may prevent motion artifacts in outcomes associated with the longer Ideal IQ technique. However, potential magnetic field inhomogeneities should be considered. The Lava Flex technique may be a faster and valid alternative to the Goutallier classification system.

中文翻译：

---

使用Lava Flex和Ideal MRI技术替代Goutallier分类系统的定量替代方法：尸体上肌的体积肌内脂肪浸润

客观的

Goutallier分类系统是对肩袖眼泪中肌内脂肪浸润进行分级的最常用方法。该分级系统的观察者间可靠性较低，并且无法为肌肉内脂肪提供定量和可重复的结果。我们确定了定量定量肌肉内脂肪的两种方法，Lava Flex和Ideal IQ MRI技术的相关性和可靠性，同时还与Goutallier方法进行了比较。

材料和方法

使用Lava Flex和Ideal IQ MRI成像协议扫描了十七具尸体肩部的棘上肌。在相同的肌肉上进行组织学分析。进行一致性，可靠性和相关性分析以比较所有结果。

结果

Lava Flex协议平均大约需要4分钟，而理想IQ需要大约11分钟才能完成。Bland–Altman分析显示了Lava Flex和理想IQ [LOA（-0.10和0.05）]之间的良好一致性，而ICC分析显示了极好的可靠性（ICC（1,1）0.948; ICC（2,1）0.947）。Lava Flex和Ideal IQ MR协议之间存在91％的相关性。组织学和Goutallier分类之间的加权Kappa分析显示，公平至中度的一致性。

讨论

Lava Flex技术占用了大约30％的采集时间，可以防止与较长的Ideal IQ技术相关的结果中出现运动伪影。但是，应考虑潜在的磁场不均匀性。Lava Flex技术可能是Goutallier分类系统的一种更快，更有效的替代方法。