Quantitative evaluation of primary lower extremity lymphedema staging using MRI: a preliminary study

Abstract

Background: The staging of primary lower extremity lymphedema (LEL) is difficult yet vital in clinical work, and magnetic resonance imaging (MRI) can be used for quantitative assessment of primary LEL due to its high resolution for soft tissues. In this study, we evaluated the value of MRI-based soft tissue area measurements for staging primary LEL.

Methods: A total of 90 consecutive patients with clinically diagnosed primary lower limb lymphoedema from January 2017 to December 2019 in Beijing Shijitan Hospital were enrolled retrospectively. Short time inversion recovery (STIR) sequence was applied to measure the total, muscle, bone, and subcutaneous areas in the upper 1/3 level of the bilateral lower calf. The difference between the affected and unaffected calf regarding the subcutaneous area was obtained, and (subcutaneous area)/(bone area) and (subcutaneous area)/(muscle area) were calculated. According to the International Society of Lymphology (ISL) clinical staging standard established in 2020, all patients were divided into stages I, II, and III, accordingly. Statistical analysis was performed to determine the validity of MRI measurements in staging LEL.

Results: There were 33 patients classified as stage I clinically, 44 patients as stage II, and 13 patients as stage III. There were significant differences in total, subcutaneous, the difference in subcutaneous area of limbs, subcutaneous/bone (S/B), and subcutaneous/muscle (S/M) between stage I and II as well as between stage I and III (P<0.001), but not between stage II and III (P=0.706, 0.329, and 0.229, respectively). A positive correlation was detected between the clinical stage and difference in subcutaneous area of limbs (rho =0.752, P<0.001), S/B (rho =0.747, P<0.001), S/M (rho =0.709, P<0.001), and subcutaneous (rho =0.723, P<0.001). For staging primary LEL, receiver operating characteristic (ROC) curves indicated that the difference in subcutaneous area of limbs had the best discrimination ability among parameters [area under the ROC curve (AUC) =0.950; 95% confidence interval (CI): 0.875-0.987; sensitivity: 95.45%; specificity: 84.85%], followed by S/B (AUC =0.930; 95% CI: 0.848-0.975; sensitivity: 77.27%; specificity: 93.94%) and S/M (AUC =0.895; 95% CI: 0.804-0.953; sensitivity: 77.27%; specificity: 90.91%). The ROC curves indicated that subcutaneous area (AUC =0.927; 95% CI: 0.844-0.974; sensitivity: 84.09%, specificity: 90.91%) and total (AUC =0.852; 95% CI: 0.753-0.923; sensitivity: 70.45%; specificity: 90.91%) also had discrimination ability between stage I and II.

Conclusions: The measurement of the soft tissue area of the calf may be used as an auxiliary method for staging primary LEL. For patients with unilateral primary LEL, the difference in subcutaneous area of limbs could be a specific indicator to distinguish clinical stage I from II.

Keywords: Lymphedema; clinical stage; lower extremity; magnetic resonance imaging (MRI); soft tissue area.

Figure 1

Flowchart of participants. Patients with primary LEL between January 2017 and December 2019 (n=537). A total of 447 patients were excluded from the analysis. The remaining 90 patients were included in this retrospective study. LEL, lower extremity lymphedema.

Figure 2

Lower extremity volume measurements by circumferential method. Primary LEL of stage II (13-year-old male, right calf). The patient’s lower extremities were measured bilaterally and the circumference of the ankle was recorded (a), lower 1/3 of the calf (b), upper 1/3 of the calf (c), knee (d), lower 1/3 of the thigh (e), upper 1/3 of the thigh (f), and the root of the thigh (g) on the affected side and the normal side, respectively, as well as the height (H) between adjacent circumference levels, and used the calculation formula to derive the segmental volumes, and added the volumes of the corresponding segments to calculate the total volume. LEL, lower extremity lymphedema.

Figure 3

MRI measurements of patients with primary LEL and STIR images at different clinical stages. (A) MRI (axial STIR sequence) measurement of the lower extremity’s total, muscle, bone, and subcutaneous area. The skin (white arrow) was defined as the outer boundary. The deep fascia (black arrow) was defined as the boundary between subcutaneous soft tissue and muscle. The boundaries of fibula and tibia were defined as the outer edge of the fibula and tibial bone cortex (thick arrow). The sum of the tibia and fibula areas was the bone area. The area within the calf skin was expressed as total area, the area between the muscle and bone boundary was expressed as muscle area, and the area between the deep fascia and the skin boundary was expressed as subcutaneous area. (B) Primary LEL of stage I (15-year-old female, right calf). (C) Primary LEL of stage II (30-year-old female, left calf). (D) Primary LEL of stage III (18-year-old female, left calf). MRI, magnetic resonance imaging; LEL, lower extremity lymphedema; STIR, short time inversion recovery.

Figure 4

Associations between soft tissue areas of lower extremities and stages of LEL. (A-E) An association between the total, subcutaneous area, the difference in subcutaneous area of limbs, S/B, S/M with LEL stage, respectively. ***P<0.001. LEL, lower extremity lymphedema; S/B, subcutaneous/bone; S/M, subcutaneous/muscle.

Figure 5

ROC-identified total, subcutaneous area, the difference in subcutaneous area of limbs, S/B, and S/M values for classifying LEL stage I vs. II. ROC, receiver operating characteristic; S/B, subcutaneous/bone; S/M, subcutaneous/muscle; LEL, lower extremity lymphedema.