Interrater reproducibility of the Myoton and durometer devices to quantify sclerotic chronic graft‐versus‐host disease
Authors: Shramana Ghosh 1, 2, Laura Baker 1, 2, Fuyao Chen 1, 2, 3, 4, Zain Khera 2, Arved Vain 2, 5, Kathy Zhang 2, Alexis Hood 2, Hayden Smith 2, Heidi Chen 6, 7, Madan Jagasia 7, Eric Tkaczyk 1, 2, 3, 7
- Dermatology and Research Services, Department of Veterans Affairs, Tennessee Valley Healthcare System, Nashville, TN, USA
- Vanderbilt Dermatology Translational Research Clinic, Department of Dermatology, Vanderbilt University Medical Center, TN, Nashville, USA
- Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA
- Medical Scientist Training Program, Yale School of Medicine, New Haven, CT, USA
- University of Tartu Institute of Physics, Tartu, Estonia
- Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA
- Vanderbilt-Ingram Cancer Center, Nashville, TN, USA
Journal: Archives of Dermatological Research - May 2023 (DOI: 10.1007/s00403-023-02626-1)
- The overall ICCs trended higher for MyotonPRO than the typical range of ICCs associated with clinical evaluation of sclerosis.
- MyotonPRO can be used for assessing cutaneous sclerosis with higher reproducibility than the modified Rodnan skin score.
Chronic graft-versus-host disease (cGVHD) is a severe complication in long-term survivors of allogeneic hematopoietic stem cell transplantation. This disease is challenging to manage clinically due to a lack of validated tools to quantitatively measure skin sclerosis.
The current gold standard for measuring skin sclerosis is the NIH Skin Score which has only moderate agreement among clinicians and experts. To more accurately assess skin sclerosis in cGVHD, the MyotonPROand durometer devices can be used to directly measure biomechanical parameters of the skin. However, the reproducibility of these devices is not known in patients with cGVHD.
To determine this reproducibility, three observers independently measured 10 anatomic sites in each of seven patients with sclerotic cGVHD using the MyotonPRO and durometer. Clinical reproducibility was measured by mean pairwise differences (U-statistic) and intraclass correlation coefficients (ICCs) with 95% confidence intervals (CIs). Mean pairwise differences, expressed in true physical units, were used to report typical errors for each anatomic site and device.
Mean pairwise differences were less than 11% of the average overall values for all five MyotonPROparameters and durometer hardness. These were lower for MyotonPRO creep (4.1%), relaxation time (4.7%), and frequency (5.1%) than decrement (9.0%), stiffness (10.4%), and durometer hardness (9.0%). MyotonPROparameters creep, relaxation time, and frequency showed promise for capturing skin biomechanics more accurately than MyotonPRO stiffness, decrement, or durometer hardness. Mean pairwise differences trended highest in the shin and volar forearm and lowest in the dorsal forearm. The interobserver ICC for overall (averaged across all measured body sites of a patient) creep (0.94; 95% CI 0.87–1.00), relaxation time (0.96; 95% CI 0.90–1.00), and frequency (0.95; 95% CI 0.88–1.00), trended higher than that for decrement (0.43; 95% CI 0.00–0.88), stiffness (0.92; 95% CI 0.81–1.00), and durometer hardness (0.82; 95% CI 0.61–1.00). Similar trends were observed in healthy participants.
These findings can help clinicians design better studies to assess therapeutic response to new cGVHD treatments and support the interpretation of future measurements.
Keywords: sclerosis, biomechanics, durometer, MyotonPRO, reproducibility, chronic cutaneous graft-versus-host disease
Overall, our results contribute towards validating the practicality of using the MyotonPRO and the durometer in future research and clinical practice. In particular, the MyotonPRO parameters of creep, relaxation time, and frequency show promise towards accurately capturing true skin biomechanics. A prospective longitudinal study following patients over time is needed to validate the use of the MyotonPRO to monitor cGVHD progression and treatment response. Inexperienced users with basic anatomy knowledge can acquire the skills to conduct these measurements with 6–8 h of expert-led, in-person training, as demonstrated by the observers KZ and AH in this study. We recommend the use of standardized measurement protocols and training materials created using the data presented here. Data interpretation will benefit from study designs that take into account the site-specific discriminatory ability and interrater reproducibility. The MPDs reported here could be used to inform whether changes in a participant’s measurements fall within the expected range of interrater differences. This would enable investigators to determine if the differences measured between two study populations might be attributed to measurement error or represent a true change in the biomechanical properties of participant’s skin. Thus, the MyotonPRO may offer a new approach to monitor disease progression and treatment responses in patients with sclerotic cGVHD.