Measuring Biomechanical Properties Using a Noninvasive Myoton Device for Lymphedema Detection and Tracking: A Pilot Study

Lymphedema is a debilitating and progressive condition that affects over 20% of patients who undergo surgical or radiation treatment for breast cancer.¹ In a normal lymphatic system, water and proteins are reabsorbed from the interstitium and returned to the arteriovenous system. Lymphedema is a pathologic state in which there is damage to the lymphatic vessels and transport mechanisms, leaving fluid to indefinitely accumulate in the extracellular space. Fluid can also move in a retrograde manner and cause tissue swelling, adipose tissue remodeling, and collagen deposition in the superficial skin lymphatics.² This “dermal backflow” causes skin layers to change its structural composition and mechanical properties.³ Following lymph node dissection and radiation therapy for breast cancer treatment, patients may develop insidious signs of lymphedema, including unilateral arm heaviness, numbness, altered physical appearance, and recurrent infections. Breast cancer–related lymphedema (BCRL) is incurable and may lead to negative psychosocial repercussions.³ Severe BCRL presents a serious economic burden long after breast cancer is cured.^4,5 Clinicians must proactively survey patients for signs and symptoms of lymphedema to initiate therapy and prevent further deterioration.

BCRL may develop years after mastectomy or radiation treatment for breast cancer, and serial testing is essential for early detection of subclinical BCRL.⁶ Traditionally, circumferential tape measurements and water displacement have been used to monitor BCRL. However, these modalities offer little information for diagnosing subclinical and mild BCRL (International Society of Lymphology [ISL] Stage 0 and Stage I, respectively).⁷ Other available diagnostic techniques include bioimpedance spectroscopy (BIS), magnetic resonance imaging (MRI), and lymphoscintigraphy. BIS noninvasively measures the extracellular fluid in the body; although it is a sensitive test, it lacks specificity.⁸ Lymphoscintigraphy and indocyanine green lymphography are effective imaging techniques used to visualize abnormal lymph flow in more severe cases. These techniques are typically only offered at large academic centers and therefore are not universally available nor affordable.⁸

Given the nonspecific and inaccessible nature of the available validated lymphedema detection methods, researchers have employed devices designed to measure changes in biomechanical properties of skin or, more specifically, skin stiffness. Such devices as the SkinFibroMeter, tonometer, and extensometer specifically measure skin stiffness in patients with lymphedema.^9-11 These devices are largely used for research, and none thus far have been approved for clinical purposes.¹² The MyotonPRO Digital Palpation Device (Myoton AS, Estonia) (Figure 1) is a noninvasive handheld device used to measure muscle and soft tissue dynamic biomechanical parameters.¹³ It delivers a short, fixed impulse to the tissue, and a signal is interpreted by the device to calculate biomechanical parameters, including dynamic stiffness (N/m), the resistance to external deformation forces; oscillation frequency (Hz), the mechanical tension in the relaxed skin; relaxation time (ms), the duration of time for the skin to restore its shape after the removal of an external force; and creep, a ratio of deformation and relaxation time.¹⁴ Although the Myoton device was developed to measure muscle, specific modifications, including the addition of a 12-mm flat disk to the probe and shortening of impulse delivery time, have allowed for a greater selection of superficial cutaneous tissue.¹⁴ Given the Myoton device’s promising results in assessing sclerotic skin conditions, such as graft versus host disease,^14-16 this study aims to determine if the device can detect changes in skin parameters between the unaffected and affected arms of patients with BCRL.

Institutional review board approval was obtained, and women diagnosed with subclinical to moderate upper extremity BCRL were recruited from the Vanderbilt-Ingram Cancer Center. Patients were excluded if their lymphedema was unrelated to breast cancer treatment or if they were less than 18 years of age. For each Myoton observation, the participant was initially positioned supine on an examination table. Using the modifications employed by Chen and coworkers, a 12-mm diameter disk was attached to the device probe and placed flat to the skin.¹⁴ The Myoton delivered a 7-ms impulse to the underlying tissue. If the device was unable to perform the measurement, the impulse time was adjusted to 10 ms. Even with an impulse duration of 15 ms, there was minimal force (0.6N); therefore, there was no potential for residual mechanical deformation or skin distortion.¹⁷ A single trained observer [LD] measured 7 anatomical sites: biceps, dorsal forearm, dorsal hand, dorsal wrist, triceps, volar forearm, and volar wrist (Figure 2). Triceps measurements were performed with the patient in the prone position. For each participant, the contralateral, unaffected arm was measured in the same fashion.

Each Myoton observation session lasted approximately 30 minutes, and an additional 30 to 60 minutes was spent with each participant to obtain circumferential tape measurements and complete Lymphedema Symptom Intensity and Distress (LSID) surveys.¹⁸ Descriptive analyses were done to calculate means, medians, and standard deviations. R version 3.6.1 (R Foundation for Statistical Computing) was used to create box and whisker plots and perform Pearson correlations, and a P value < .05 was determined as significant. This pilot study was not designed with hypothesis testing, and therefore additional statistical analyses were not performed.

Eleven female participants with unilateral BCRL ranging from ISL Stage 0 to II were enrolled. The mean age and body mass index (BMI) of the participants were 66 years and 32.8 kg/m², respectively. The mean time since lymphedema diagnosis was 16 years (Table 1, Table 2). Generally, mild skin involvement was observed at all sites of the affected arm in all 11 participants. Data collected from the LSID-Arm survey revealed that 2 participants reported symptoms of hardness in their arms, giving a symptom prevalence of 18.2% (Table 3).

Averages for the Myoton device measurements of stiffness, frequency, creep, and relaxation time at the 7 sites in the affected arm of each participant were compared with the averages of the 7 measurements obtained in each participant’s contralateral unaffected arm (Figure 3). To normalize the data, ratios were used to compare the averages of measurements from all the affected arms (BCRL) with those from the unaffected arms (N = 11) (Figure 4, Table 1).

The stiffness and frequency measurements in most affected arms were decreased compared with the contralateral control arms (ratio < 1). The mean stiffness of all arms affected by BCRL was 533.6 ± 305.9 N/m while the mean stiffness in the unaffected arms was 590 ± 381.3. The median ratio of skin stiffness comparing the affected to unaffected arms was 0.91 (interquartile range [IQR] 0.78-1.03). The mean frequencies of the affected and unaffected arms were 21.3 ± 8.3 Hz and 22.8 ± 9.3, respectively. The median frequency ratio was 0.94 (IQR 0.89-1.0).

In comparison, the average creep and relaxation time measurements were found to be greater in the affected arms than in the unaffected contralateral arms (ratios >1). The average relaxation time in the affected arms was 18.2 ± 8.3 ms while the unaffected arms measured 17.1 ± 8.3 ms on average. The median relaxation time ratio was 1.07 (IQR 1.02-1.14). The mean creep measurements in the affected and unaffected arms were 1.9 ± 0.8 and 1.8 ± 0.8, respectively, and the median creep ratio was 1.06 (IQR 1.03-1.16). A positive correlation was observed between the skin stiffness in the affected and unaffected arms and between the skin frequency in the affected and unaffected arms (Pearson correlation coefficient r = 0.73 and r = 0.74, respectively) (Figure 5). The correlation r values of the stiffness ratios to circumference ratios, frequency ratios to circumference ratios, relaxation time ratios to circumference ratios, and creep ratios to circumference ratios were 0.310, 0.378, 0.376, and 0.380, respectively.

Although originally engineered to test muscle viscoelastic responses, the Myoton device has produced promising results when analyzing skin biomechanical parameters.^13,14,16 Our results suggest that the device can infer differences in the affected and unaffected skin in the arms of patients with BCRL. One unexpected finding was that the skin stiffness was reduced in the affected arms compared with that in the unaffected arms. Classically, histological and immunohistochemical lymphedema research has supported the notion that lymphedema causes increased skin stiffness due to increased collagen fiber deposition and fibrosis in the edematous subcutaneous and dermal layers.^2,3,9,11 Albeit tonometry devices are not recommended for diagnosis,¹² clinical trials have utilized them to measure the presence of fibrosis. Tonometer-measured values are typically higher in the lymphedematous limb because of greater fibrosis.¹⁰

Other skin stiffness research devices (eg, SkinFibroMeter and extensometer) have corroborated this concept of increased skin stiffness as well.^9,11 However, it should be noted that, in the present study, 82% of participants reported no symptoms of hardness in their affected arms. Additionally, Coutts et al¹¹ found skin stiffness to be dependent on the direction of skin measured and that, when measured “along” the arm, it decreases with the duration of the lymphedema diagnosis. Our contradictory findings suggest that the Myoton device may be unintentionally capturing excess adipose tissue and lymph fluid concentrated within the subcutis layer, a layer that is known to swell due to its low stiffness.¹¹

Still, the pathophysiology of chronic lymphedema is not completely understood,² and skin undergoes many fibrosclerotic changes as a result of the disease pathology, treatments, and other external factors. Interestingly, 4 of the 11 participants did not routinely wear compression garments; of these 4, 2 participants were found to have increased stiffness and frequency measurements in their affected arms (Table 1). Zaleska et al¹⁰ found lower tissue tonometry values after pneumatic compression of the affected limb, thus there may be a correlation between compression garment use and our device findings of decreased skin stiffness. However, given the small sample size, we could not predict whether this occurred by chance.

To our knowledge, only one other study has published findings using the Myoton parameters (stiffness, oscillation frequency, relaxation time, and creep) to evaluate tissue in BCRL. Naczk et al¹⁹ found a significant correlation between the L-Dex index and the Myoton properties of the biceps brachii. Similar to the present study, the mean stiffness and oscillation frequency (tone) were decreased in the affected arms of patients with BCRL when compared with the unaffected control arms. Additionally, the mean relaxation time and creep in the affected limbs were generally higher in the affected limbs of patients with BCRL.¹⁹ The authors postulated that the decreased stiffness and increased relaxation time and creep may be explained by lymph accumulation, which deforms the tissue and prolongs the skin’s ability to return to its natural state after pressure is applied. Furthermore, decreased oscillation frequency, or the mechanical tension in the relaxed skin, may be explained by the decreased muscle mass due to muscle atrophy after lymphatic infiltration and extended underuse.¹¹

This pilot study is limited by its relatively small size. It is possible that the duration of lymphedema may alter skin viscoelastic properties; therefore, larger longitudinal studies are needed for further comparisons and subsequent analyses to make stronger conclusions. Hypothesis testing is needed to better understand the Myoton device’s discriminatory abilities. Lastly, the device has demonstrated its ability to differentiate muscle and skin properties; however, future studies should determine the depth of the device signal and the reproducibility of measurements to enable longitudinal monitoring of disease progression.

Earlier detection of BCRL could optimize therapeutic and surgical interventions and lessen the severity of the consequent arm heaviness, disability, and negative psychosocial impact. These preliminary results show that the Myoton device, when used in this patient population, can depict some differences and changes in stiffness, frequency, relaxation time, and creep in lymphedematous limbs when compared with the contralateral, unaffected sides. Larger longitudinal studies are still needed to test the device’s reliability and accuracy in this patient population. Future studies should control for specific variables, including age, compression sleeve use, physical therapy compliance, and lymphedema staging.

Affiliations: ¹Department of Plastic Surgery, Vanderbilt University Medical Center, Nashville, TN; ²Department of Veterans Affairs and Vanderbilt Dermatology Translational Research Clinic (VDTRC.org), Nashville, TN; ³Vanderbilt University School of Medicine, Nashville, TN; ⁴Vanderbilt-Ingram Cancer Center, Nashville, TN; ⁵Division of Plastic and Reconstructive Surgery, Mayo Clinic Jacksonville, FL

Funding: This work is partially supported by Career Development Award Number IK2 CX001785 from the United States Department of Veterans Affairs Clinical Science R&D (CSRD) Service to Eric Tkaczyk.

Ethics: Institutional review board approval was obtained for this study.

Disclosures: The authors disclose no relevant financial or nonfinancial interests.