Introduction:

Spontaneous mechanical activities of the human musculature (SMAM¹) have shown a high correlation to muscular electrical activity² as well as anatomical fiber orientation³. SMAMs can be quantitatively assessed by diffusion-weighted imaging (DWI). To reveal temporal changes for a progressive quantification of the same subjects, as well as to distinguish between healthy population and subgroups with different metabolic or neuromuscular condition, the longitudinal behavior of spontaneous activities is of interest to ensure reliable grading and to reduce bias by systematic errors, e.g., time point of measurement during the day. Therefore, a data base with longitudinal measurements from a small cohort with 13 subjects is evaluated for the first time in this work. Furthermore, short-term variation was analyzed in order to assess the influence of the measurement time point on activity distribution.

Methods:

A data base of 46 individual DWI time-series measurements from 13 subjects (age: 36.5±11.5 years, BMI: 26.8±2.9 kg/m²) were retrospectively evaluated. All measurements were conducted on 3 T MR systems (MAGNETOM Trio, Skyra, Prisma^fit, Vida, Siemens Healthcare, Erlangen, Germany) with a 15-channel Tx/Rx knee-coil or body-array coils. Series of 500 repetitions of transverse images were recorded at the maximum diameter of the right calf or in the middle position of the right thigh in supine position with a prototype diffusion-weighted stimulated-echo EPI (STE-DWI) sequence or a diffusion-weighted Stejskal-Tanner spin-echo EPI (SE-DWI) sequence. Measurement parameters: matrix size: 64-80 × 64-80, FoV: 190-240 × 190-240 mm², TE: 26 ms (STE-DWI), 53 ms (SE-DWI), TR: 500 ms, slice-thickness: 6 mm, b-value: 100 s/mm², SPAIR for fat suppression and mixing time TM: 145 ms (STE-DWI). Subjects were sorted regarding date of measurement and time point in three different subsets. These subsets are given in Table 1: Short-term: Three subjects with DWI measurements on the same day (in the morning, noon and evening). Medium-term: 12 subjects with DWI measurements within one year, whereas 7 subjects were from a diabetes prevention study with an exercise training intervention between both measurements. The training program lasted 8 weeks and consisted of three supervised exercise sessions per week. Each training session consisted of 30 min of bicycle ergometer exercise and 30 min walking on a treadmill at 80% of VO2_peak. Long-term: Four subjects with DWI measurements over 23-62 months. Care was taken that all subjects rest before DWI measurement and that imaging parameters which can influence the sensitivity of the MR sequence to image spontaneous activities, e.g., excitation scheme, b-value and mixing time, were in consensus within one subject. Data processing: First, all DWI within a time-series were co-registered. Therefore, a common space reference image was calculated based on principle component analysis. Subsequently, all images were registered on this reference image by a Local All-Pass registration^4,5. SMAMs in DWI were evaluated by a semi-automated graph-based segmentation approach⁶. Evaluation: Gross movements of the musculature were manually discarded before further evaluation. Number of events and event count maps (ECM) were assessed to analyze the temporal behavior of spontaneous activities.

Results & Discussion:

Intra-day spontaneous activity distribution is evaluated for three subjects in Fig. 1A. Only minor variability is revealed for Subject #1 and #3, whereas one time point of Subject #2 shows increased activity. ECMs in Fig. 1B show that the same muscular regions are active in the morning, noon as well as evening. Mean time-difference between DWI measurements for the medium-term evaluation was 91±58 days. Number of SMAM-affected DWIs and ECMs are depicted in Fig. 2 for healthy subjects and exercise program attendees (dashed). Activity seems not to be influenced by the exercise program in this special population. Evaluation of the long-term analysis is depicted in Fig. 3. Subject #2 shows an increase in activity, but spontaneous activity was already high four years ago in comparison to other subjects. Moreover, spontaneous activity is present in the same muscular regions over years.

Conclusion:

For the first time, long-term behavior of spontaneous activity assessed by DWI was evaluated. This work reveals only moderate variation of the frequency of spontaneous muscular activities for most subjects over a time-period from several hours to years. Hence, quantification of spontaneous activity can be potentially utilized as additional information in progressive and longitudinal studies. Moreover, spontaneous activity seems not to be correlated to the MR system, since different systems were utilized for image acquisition. Due to the small amount of subjects, larger cohort studies based on this work should be conducted to get detailed insight in the long-term behavior and influencing factors.

Acknowledgements

We thank Stemmer, A., Siemens Healthcare GmbH, for his valuable technical support on this project. This work was supported and funded by the German Research Foundation (DFG) under Grants SCHI 498/11‐1 and YA 28/16‐1 and in part by a grant (01GI0925) from the German Federal Ministry of Education and Research (BMBF) to the German Center for Diabetes Research (DZD e.V.).