The value of nonenhanced magnetic resonance imaging (MRI) in the grading of primary lower extremity lymphedema

Objective: The International Society of Lymphology (ISL) proposed a grading standard for lymphedema in 2020 based on the percent increase in the volume of the affected limb compared to that of the healthy limb. However, this method is cumbersome and time-consuming to measure and calculate, and a standardized formula across different institutions is not available. Therefore, the aim of this study was to investigate the value of nonenhanced MRI for grading primary lower extremity lymphedema (PLEL).

Methods: This retrospective study included 124 consecutive patients with unilateral PLEL from 2021 to 2023. All patients were categorized into three groups, mild(n=43), moderate(n=41), and severe(n=40), according to the 2020 ISL grading standard. From the lymphedema involvement range (vertical range: whole lower extremity, only thigh, only calf and ankle; transversal range: ≤25% of the cross section, 26%-50%, 51-75%, >75%), MRI signs of lymphedema (parallel lines sign, grid sign, honeycomb sign, band sign, crescent sign, lymphatic lake sign, and nebula sign), and lymphedema measurements (total diameter, total circumference and total area of the affected limb; diameter and area of the bone, muscle, subcutaneous fat and subcutaneous soft tissues on the affected limb; circumference of the bone and muscle on the affected limb; thickness of skin; thickness of band sign; thickness of crescent sign) were recorded and statistically analysed in the three groups of patients.

Results: The statistically significant differences in the indicators among the three groups were as follows: vertical and transversal ranges of lymphedema, parallel lines sign, grid sign, honeycomb sign, band sign, crescent sign, and lymphatic lake sign, total diameter, total circumference, total area, diameter and area of the subcutaneous fat, diameter and area of the subcutaneous soft tissues, thickness of skin, thickness of band sign and crescent sign (P<0.05). The receiver operating characteristic (ROC) curve showed that the highest area under the curve (AUC) for each parameter for identifying patients in the mild and nonmild (including moderate and severe) groups was in the following order: diameter of the subcutaneous fat> area of the subcutaneous fat> thickness of the skin(P<0.05). The ROC curve showed that the highest AUC for each parameter used to identify patients in the severe and nonsevere (including mild and moderate) groups was in the following order: diameter of the subcutaneous fat > area of the subcutaneous fat > thickness of the crescent sign.

Conclusions: (i) The parallel lines sign is a characteristic indicator for diagnosing patients with a mild disease, the grid sign is a characteristic indicator for diagnosing patients with a moderate disease, the lymphatic lake sign and crescent sign are characteristic indicators for diagnosing patients with a severe disease, and the honeycomb sign and band sign are characteristic indicators for diagnosing patients with a moderate-severe disease. (ii) The thickness of the skin, band sign and crescent sign gradually increased with increasing disease severity. (iii) The efficacy of the diameter and area of subcutaneous fat for PLEL grading is optimal. (iv) Nonenhanced MRI can be a better and standardized tool for grading PLEL.