Assessment of calf muscle constitution in chronic Achilles tendon disease using Dixon-based MRI

Abstract

Objectives: To assess calf muscle constitution in chronic Achilles tendon disease (ATD) using two-point Dixon-based MRI (2pt-MRI_DIXON).

Materials and methods: This retrospective study analyzed 91 patients (36 females; 57.0 ± 14.4 years) with midportion or insertional chronic ATD who underwent clinical MRI of the Achilles tendon (AT), including 2pt-MRI_DIXON for quantitative assessment of calf muscle fat content (MFC). Additionally, two radiologists qualitatively assessed MFC, AT quality, and co-pathologies. 2pt-MRI_DIXON-derived fat fractions (FF) were related to patients' demographics and qualitative imaging findings.

Results: The overall mean FF derived from 2pt-MRI_DIXON of the triceps surae muscle was 11.2 ± 9.3%. Comparing midportion and insertional ATD, there was no significant difference regarding fatty muscle infiltration assessed with 2pt-MRI_DIXON (P ≥ .47) or qualitative grading (P ≥ .059). More severe AT thickening (11 vs.9 mm, P < .001) and complete tears (29 vs. 9%, P = .025) were significantly more common in midportion ATD, while partial tears were significantly more frequent in insertional ATD (55 vs. 31%, P = .027). Soleus muscle edema was more prevalent in midportion than insertional ATD (40 vs. 9%, P = .002). In contrast, insertional ATD more commonly featured bone marrow edema (61 vs. 2%), Haglund's deformity (67 vs. 0%), and retrocalcaneal bursitis (82 vs. 43%) (P ≤ .002). Significant correlations (P ≤ .001) were demonstrated between FF, AT diameter, age (both in midportion and insertional ATD), and body mass index (in midportion ATD only) (ρ range = 0.53-0.61).

Conclusion: In chronic ATD, calf MFC was statistically equivalent (approximately 11%), irrespective of the localization of tendon damage. More severe tendon thickening and complete tears were more common in midportion ATD, and, vice versa, partial AT tears were significantly more frequent in insertional ATD.

Keywords: Achilles tendon disease; Fat fraction quantification; Magnetic resonance imaging; Muscle; Two-point Dixon technique.

Fig. 1

Flowchart illustrating the patient selection process and the study setup. Of 153 potentially eligible patients who underwent clinical MRI of the Achilles tendon (AT) due to chronic Achilles tendon disorders (ATD), 62 were excluded during the selection process. This resulted in a study sample of 91 patients with chronic ATD, of whom 58 had midportion ATD, and 33 had insertional ATD

Fig. 2

Quantitative volumetric assessment of calf muscle fat content (MFC) using the example of a 68-year-old female (left side). Outlining of calf muscles was done in conjunction with the corresponding transversal T1-weighted sequence (A). The reconstructed two-point Dixon gradient-echo (GRE) MR images were uploaded into ITK-Snap (B). Next, the medial gastrocnemius (MG), lateral gastrocnemius (LG), and soleus muscles were manually outlined on all slices (C). After that, automated three-dimensional (3D) reconstructions were generated with an output of the mean volumetric fat fraction (FF) and whole muscle volume (WMV) for each muscle separately. In this case, the MG muscle demonstrated a FF of 10.3% (WMV = 145.8 cm³), the LG muscle of 8.6% (WMV of 85.3 cm³), and the soleus muscle of 10.7% (WMV = 211.2 cm³). These segmentations (MG, LG, and soleus muscles) were summed up to determine the FF and WMV of the triceps surae muscle (D). In this example, the FF of the triceps surae muscle was calculated with 10.2% (WMV = 442.3cm³). E, F The corresponding 3D reconstruction images

Fig. 3

Three cases of midportion (MP) Achilles tendon disease (ATD) with increasing severity of structural Achilles tendon (AT) damage as demonstrated on sagittal fat-saturated (fs) proton-density (PD)-weighted images (A–C) and corresponding transversal T1-weighted images of calf muscles at the level of the middle third of the lower leg (D–F) representatively illustrating fatty muscle infiltration visually assessed by using the Goutallier classification in comparison to quantitatively determined muscle fat content (MFC). Image A shows a case of left-sided MP ATP with spindle-shaped thickening of the AT at the level of the middle third of the tendon (arrow) in a 68-year-old female. Minimal hyperintense signal alterations within the tendon substance can be seen at the proximal and distal ends of the spindle-shaped thickening (dashed arrows). In B (right AT of a 58-year-old male), there is even more severe spindle-shaped thickening of the AT (arrow) and a partial tear of the AT at the middle third of the tendon. In contrast, a complete AT rupture in a 62-year-old male (right side) is visible in C, where one can depict a discontinuity of the AT at the level of the middle tendon third (arrow). The distal tendon stump has comparatively smooth edges (circle). In the first case (D), visually assessed fatty infiltration of the medial (MG) and lateral gastrocnemius (LG) muscles was rated as a grade 1, whereas that of the soleus muscle was rated as a grade 2. Fatty muscle infiltration of the LG muscle in the second case (E) was assessed as a grade 1, while the MG and soleus muscle fatty infiltration were rated as grade 2. In the third case (F), the degree of fatty infiltration of the MG and LG muscles was evaluated as a grade 2, while the soleus muscle was assessed as a grade 3. Quantitative fat fraction values were 10.3%, 8.6%, and 10.7% (MG, LG, soleus muscle) in the first case (A, D); 11.1%, 9.4%, and 12.2% in the second case (B, E); and 14.1%, 9.3%, and 13.9% in the third case (C, F). These case examples underline the inferior discriminatory power of qualitative compared to quantitative MFC determination

Fig. 4

Three cases of insertional (IN) Achilles tendon disease (ATD) with increasing severity of structural Achilles tendon (AT) damage as demonstrated on sagittal fat-saturated (fs) proton-density (PD)-weighted images (A–C) and corresponding transversal T1-weighted images of calf muscles at the level of the middle third of the lower leg (D–F) representatively illustrating fatty muscle infiltration visually assessed by using the Goutallier classification [26] in comparison to quantitatively determined muscle fat content (MFC). Image A depicts IN ATD in a 45-year-old female (right side), which is characterized by tendon thickening and minimal intratendinous hyperintensities on fluid-sensitive images (arrow). There is severe inflammation of the paratendinous soft tissues with significant Kager’s fat pad edema (asterisk) and retrocalcaneal bursitis (dashed arrow). Image B illustrates IN ATD in a 54-year-old female (left side) with a partial rupture at the tendon insertion (arrow, asterisk), a Haglund’s deformity of the calcaneus and retrocalcaneal bursitis (circle), but no Kager’s fat pad edema. A case of IN ATD with complete rupture of the AT in a 72-year-old male (right side) is shown in image (C). There is discontinuity of the AT directly at its insertion (arrow, yellow circle). In addition, there is mild Haglund’s deformity with edema in the posterosuperior part of the calcaneus (green circle) and extensive fluid in the soft tissues. In the first patient example (D), visually assessed fatty infiltration of the medial (MG) and lateral gastrocnemius (LG) muscles was rated as a grade 1, whereas that of the soleus muscle was rated as a grade 2. In the second case (E), visual assessment of fatty muscle infiltration was assessed as a grade 1 for all three muscles. In the third case (F), the degree of fatty infiltration of the LG and soleus muscles was evaluated as a grade 2, while the MG muscle was assessed as a grade 3. Quantitative fat fraction values were 4,7%, 4.6%, and 6.9% (MG, LG, soleus muscle) in the first case (A, D); 9.3%, 9.2%, and 9.5% in the second case (B, E); and 21.1%, 11.0%, and 13.6% in the third case (C, F). Analogous to Fig. 3, these case examples underline the superiority of quantitative over qualitative MFC determination

Fig. 5

Assessment of co-pathologies in 58 patients with midportion (MP) vs. 33 patients with insertional (IN) Achilles tendon disease (ATD). Soleus muscle edema was significantly more common in patients with MP than IN ATD (P = .002). Haglund’s deformity, bone marrow edema in the posterosuperior calcaneus, and retrocalcaneal bursitis (all P < .001) were significantly more often observed in patients with IN than MP ATD. No significant differences between the two subtypes of ATD were found for edema in the MG or LG muscle and Kager’s fat pad edema