Journal Logo


Neuromuscular Electrical Stimulation Preserves Leg Lean Mass in Geriatric Patients


Author Information
Medicine & Science in Sports & Exercise: April 2020 - Volume 52 - Issue 4 - p 773-784
doi: 10.1249/MSS.0000000000002191


Age-related loss of muscle mass is associated with poor mobility, loss of independency, increased hospitalization, and increased mortality (1). In the present study, the term “geriatric” is used for elderly (≥65 yr) individuals admitted to a geriatric hospital ward, whereas the term “elderly” is used for healthy nonhospitalized individuals (≥60 yr). The number of nonconsecutive days spent in a hospital during a year is associated with loss of muscle mass and strength in elderly individuals (2), and it has therefore been suggested that the age-related loss of muscle mass is accelerated during shorter periods of illness or hospitalization followed by incomplete recovery (3,4). Whether hospitalization associated loss of muscle mass actually occurs during the hospital stay in patients admitted to a geriatric ward has, however, only been sparsely investigated (5,6). In healthy elderly individuals, it has been shown that aging is negatively associated with recovery of muscle mass after an experimental model of disuse atrophy (7). These authors furthermore noted a relationship between muscle regrowth and satellite cell (SC) proliferation during rehabilitation, with a reduced SC response in healthy elderly compared with healthy young subjects (7). The SCs are essential for muscle fiber regeneration and repair and are the source of new myonuclei during muscle fiber hypertrophy (8); thus, preservation and activation of SCs during incidents of short-term atrophy may be beneficial for subsequent recovery. It is, however, debated whether SCs are lost during short-term muscle fiber atrophy (9,10).

Several factors may contribute to loss of muscle mass during hospitalization in geriatric patients, that is, nutritional status, activity level, and systemic factors related to their medical condition (3,4,11). In healthy elderly individuals, various models of inactivity can induce muscle atrophy (12–14), but this can be prevented with bouts of heavy resistance training (14) or neuromuscular electrical stimulation (E-Stim) (15). In contrast, geriatric patients are not completely inactive but are often characterized by a low activity level (16). In addition, endocrine and inflammatory responses in these patients may result in further elevated muscle protein breakdown (3,4,11), potentially accelerating loss of muscle mass compared with models of disuse atrophy in healthy individuals. The literature is sparse regarding both loss of muscle mass (5,6) and the effect of daily bouts of muscle activation during hospitalization in geriatric patients (5), and to our knowledge, this has not been investigated at the cellular level in muscle biopsies. We have previously demonstrated a positive effect on leg lean mass with a unilateral heavy resistance training model in geriatric patients (5), but in that study, we did not observe a decline in lean mass in the untrained leg. We speculated if there was a cross-education effect from the unilateral heavy resistance training intervention, as lean mass in an immobilized limb may be preserved by resistance training of the nonimmobilized limb (17). In contrast, we are not aware of studies reporting a cross-education effect on muscle mass in the nonstimulated limb after unilateral neuromuscular E-Stim. E-Stim may therefore serve as a more suitable unilateral study design in order to examine both the loss of leg lean mass during a hospital stay and the counteracting effects of daily bouts of muscle activation. Furthermore, heavy resistance training may have some limitations in a geriatric patient group, particularly in those with poor mobility and motivation or in patients who are tired, dizzy, or in pain. In contrast, E-Stim can be applied while the patients are lying in their bed, and can potentially induce forceful muscle contractions if tolerated by the patients. Although E-Stim has shown promising effects on leg lean mass during bed rest in healthy elderly (15) and on muscle fiber size in critically ill comatose patients (18), it has not been investigated whether E-Stim is an effective method to preserve muscle mass in geriatric patients.

The aim of the present study was therefore to investigate the effect of a hospital stay either with or without daily sessions of E-Stim–induced muscle contractions on changes in muscle mass, muscle fiber size, SC content, and muscle gene expression levels in geriatric patients. It was hypothesized that E-Stim would preserve lean mass, muscle fiber size, and SC content; increase the number of activated SCs; downregulate genes associated with muscle proteolysis; and upregulate genes associated with muscle protein synthesis and cellular stress.


Ethical Approval

The study was approved by the Research Ethics Committees of the Capital Region of Denmark (H-15005016) and conformed to the standards set by the Declaration of Helsinki, except for registration in a database. All patients signed a written informed consent agreement upon inclusion to the study.


The study was conducted in a 31-bed geriatric ward at Bispebjerg Hospital, Capital Region of Denmark, from August 2015 to July 2016. Patients admitted to this geriatric ward were ≥65 yr old. During this period, 805 patients were admitted to the department and screened for eligibility to participate in the study. After approval by the responsible medical doctor at the department, Danish-speaking, nonterminal, nonisolated patients with an expected length of stay of ≥7 d were included if they were cognitively well functioning (no history of dementia or Alzheimer’s disease), medically stable enough to participate, physically able to complete the test battery, did not have a medical condition known to accelerate loss of muscle mass (terminal cancer, congestive heart failure, severe chronic obstructive pulmonary disease, HIV/AIDS), had similar function and size of the lower limbs, and did not have a pacemaker/implantable cardioverter–defibrillator. Patients who were not eligible to, or declined to, participate in the muscle biopsy procedure were either offered to participate in a heavy resistance training study in the same department (5) or included in the present study without the muscle biopsy procedure. A total of 17 patients were included in the study (6 males, 11 females). Four female patients did not complete the study because the test battery was too exhausting (n = 2) or because of worsening of their medical condition not related to the study (n = 2). Therefore, a total of 13 patients (6 men/7 women) completed the intervention, and muscle biopsies were collected in 9 of these patients (5 men/4 women). Two patients were admitted directly to the geriatric ward, whereas 11 patients were initially admitted to an acute care facility for 1 d (n = 6) or 2 d (n = 5), before being transferred to the geriatric ward. Anthropometric data of the included patients are presented in Table 1. Most of the patients were admitted with more than one diagnosis, with the most frequently occurring being fall incident (n = 4), lower back pain (n = 5), infection (n = 4), pneumonia (n = 3), dehydration (n = 2), and dizziness (n = 2).

Descriptive data of the patients at the time of admission.

Study Design

At the first day of admission to the department (day 1), eligible patients received oral and written information of the study and signed an informed consent the following day after approval by the responsible medical doctor. A test battery was completed at admission (Pre; day 2) and discharge (Post; days 8–10). The test order was consistently repeated, starting with muscle scans followed by muscle biopsies and finally tests of muscle function and general function. The intervention consisted of a daily E-Stim session of the m. vastus lateralis (VL) and m. vastus medialis (VM) muscles of one leg (E-Stim), whereas the other leg served as a control leg (CON). E-Stim was randomly assigned to the strongest or weakest leg, starting with familiarization to the E-Stim protocol at the end of the pretest day. The following days (7 d·wk−1), one daily E-Stim session was completed, until and including the day before the posttest (day of discharge or no later than day 10 of the hospital stay). No adverse effects of the E-Stim protocol were observed, and no patients dropped out of the study because of the E-Stim sessions. There was a 100% compliance to the E-Stim protocol, with an average of 6.6 ± 1.0 d and E-Stim sessions between the pretest and posttest. Activity level was recorded as the percent time spent inactive (lying down/sitting down) during the hospital stay with an ActivPAL accelerometer (PAL Technologies, Glasgow, Scotland) attached to the front of the thigh. A reliable step count could not be recorded in these patients.

E-Stim Intervention

One daily 30-min E-Stim session was applied to the knee extensor muscles (VL and VM) of the E-Stim leg, with a CefarCompex mi-Theta 600 Muscle stimulator (DJO Nordic, Malmoe, Sweden). A pillow was placed under the knee of the stimulated leg, ensuring a slightly bended knee (50° knee angle, 0° = fully straight leg). A 10 × 5-cm self-adhesive electrode was placed on the midline of the anterior part of the thigh, 5 cm distal to the inguinal crease. One 5 × 5-cm electrode was placed at the distal part of the VL and VM, respectively, after a motor point search, and the position of the electrodes was marked for accurate relocation on following days. The stimulation protocol was adopted from two recent studies showing positive effect of E-Stim during immobilization (19) and in comatose patients (18). Briefly, the protocol consisted of a 5-min warm-up/cool-down sequence (5 Hz, 250 μs) and a 30-min stimulation sequence (5-s on/10-s off, 100 Hz, 400 μs, 0–120 mA) with a 0.75-s rise/fall and a 3.5-s contraction phase. The patients were instructed not to co-contract the muscles during the stimulation. The stimulation intensity was gradually increased during every session and from session to session, as tolerated by the patient. The stimulation intensity of the VL in the first session started at 42 ± 19 mA (mean ± SD) and increased gradually to 61 ± 21 mA at the end of the session. Stimulation intensity in the last session started at 68 ± 26 mA and increased gradually to 89 ± 24 mA at the end of the last session. When the stimulation resulted in a heel lift from the mattress, the leg was held down to the mattress by placing a hand on the ankle or with a strap during forceful contractions.


Functional performance

Whole-body functional testing included a test for mobility (DeMorton Mobility Index; DEMMI), lower limb strength and endurance (30-s chair stand test; 30-s CST), and gait speed (4-m gait speed test; 4-m GST), performed according to the guidelines (20–22).

Muscle power and muscle strength

Lower limb extension power (muscle power) was tested in a Nottingham PowerRig (Queen’s Medical Centre, Nottingham University, United Kingdom). After familiarization, five maximal attempts were performed with the same leg, followed by additional trials until the maximal score was followed by two lower attempts.

Maximal isometric knee extension torque (maximal voluntary contraction; MVC) was measured in a customized chair as described by Norheim et al. (5). After a warm-up trial, a total of three MVCs (6 s per trial) were performed for each leg, alternating between the legs with 2-min rest between attempts with the same leg. The maximal force (N) for each leg was recorded from the best trial, and the MVC (N·m) was calculated using the moment arm (m) from the dynamometer to the rotational axis of the knee.

Handgrip strength was measured with a Jamar® hydraulic hand dynamometer at admission only, serving as a descriptive measure of the patients. The patients performed a minimum of three attempts with both hands, until no further improvements were registered.

Lean mass (dual-energy x-ray absorptiometry)

A whole-body dual-energy x-ray absorptiometry (DEXA) scan was performed in a Lunar DPX-IQ DEXA scanner (GE Healthcare, Chalfont St. Giles, United Kingdom) and analyzed with the standard software package (Lunar iDXA Forma enCORE vs.15) by an investigator blinded to patient ID, time, and treatment. Patients were positioned in the scanner with 10-cm distance between the heels to separate the lower limbs in the subsequent analysis. Lean mass was defined as soft tissue without fat mass and bone. Segmentation of body parts was performed according to the manufacturer’s guidelines, and leg lean mass data were extracted from this analysis. The skeletal muscle index (SMI = appendicular lean mass/height2) was calculated and compared with cutoff levels for sarcopenia (1) for men (<7.26 kg·m−2) and women (<5.45 kg·m−2). Midthigh region lean mass was defined as lean mass in a 4-cm region at the midthigh level with a method slightly modified by Norheim et al. (5). The midthigh region of the thigh was included because it may be the region where the largest relative changes in muscle size after inactivity occurs (23), and because it corresponds to the site where muscle biopsies were collected. Briefly, with the Lunar iDXA software, a stack of 4-cm-thick boxes was placed over the thigh, starting at the most distal part of the lateral condyle of the femur. Femur length was measured in the Lunar iDXA software, and 50% femur length was defined as half the length from the most proximal point of the greater trochanter of the femur to the tibia plateau. The box covering the position corresponding to 50% femur length was included for analysis of midthigh region lean mass.

Muscle thickness (ultrasound)

Ultrasound (US) scans of the knee extensor muscle thickness (VL + m.vastus intermedius (VI)) in the transversal plane of the thigh at 50% femur length were included to complement the DEXA scans. A detailed description of the method for recording the US images has been published elsewhere (24). Briefly, the thickness of the VL and VI was measured by an experienced operator with B-mode ultra sound (GE Medical Systems; LogiQe), while the patients were seated on a chair with 90° flexion in the hip and knee joint. The deep aponeurosis separating VL and VI often had a wavy shape, preventing a reliable measurement of VL thickness; therefore, the combined thickness of VL + VI was measured. A total of seven images were recorded for each measurement, the images were analyzed five times with ImageJ (version 1.51n; National Institute of Health, Bethesda, MD), and the highest and lowest value was discarded before calculating the mean VL + VI thickness for each image. The two highest and two lowest scoring images were then discarded before VL + VI thickness was calculated as the mean of the remaining three images.

Muscle biopsies

A total of four muscle biopsies were taken, one Pre biopsy and one Post biopsy from the VL in both legs. Biopsies were taken with 2–3 cm between incision sites, randomized with respect to proximal/distal position between Pre and Post biopsies. Biopsies were taken after muscle scans, and the procedure was followed by 1–2 h of rest before the functional tests. The Post biopsies were taken approximately 20 h after the last E-Stim session because of logistical reasons. Briefly, at the morning conferences, it was decided by the medical team if a patient was discharged from the department on the same day, and therefore, it was not possible to plan ahead before the posttest. The E-Stim sessions were performed every day in the afternoon, being the time of day where it was most likely to get access to the patients, resulting in ~20 h from the last E-Stim to the posttest biopsies. After making an incision in the skin, biopsies were collected with the Bergström percutaneous needle biopsy technique and manual suction under local anesthetic (1% lidocaine). The specimen was carefully aligned, embedded in Tissue-Tek®, and frozen in isopentane precooled by liquid nitrogen. The embedded samples were stored at −80°C until all biopsies were collected. Ten-micrometer-thick sections were then cut in a cryostat at −20°C, placed on glass slides, and stored at −80°C.


Three staining protocols were performed on separate glass slides for the analysis of staining: 1) muscle fiber cross-sectional area (CSA) and fiber type, 2) SC and fiber type, and 3) proliferating SC. Detailed description of the staining protocols can be found in Supplemental Digital Content (Document, Supplemental Digital Content 1, Methods,; primary and secondary antibodies are listed in Supplemental Table S1 (Supplemental Digital Content 2, primary and secondary antibodies for immunohistochemistry,

Image acquisition and analysis

All image analysis was performed in ImageJ (version 1.51n; National Institute of Health) by an investigator blinded to subject ID, treatment, and time point. Fiber CSA was analyzed with a semiautomatic macro in ImageJ, according to methods published elsewhere (25). A detailed description of all image analyses can be found in Supplemental Digital Content File (Document, Supplemental Digital Content 1, Methods,

On average, 249 ± 130 (range, 89–696) type I fibers and 235 ± 145 (range, 15–669) type II fibers were analyzed for muscle fiber CSA. SCs were analyzed in 518 ± 292 (range, 109–1430) type I fibers and 550 ± 454 (range, 30–2213) type II fibers. One patient had very few type II fibers in the Post biopsies, with 15/30 (Post CON) and 52/60 (Post E-Stim) type II fibers included in the CSA/SC analysis, respectively. In addition, fiber CSA was analyzed in 89 type I fibers in one biopsy and in 67 type II fibers in one biopsy, whereas all other fiber CSA and SC analyses included more than 100 fibers for each fiber type. All regions with muscle tissue (7.2 ± 2.9 mm2), excluding large bands of connective tissue, were analyzed for proliferating cells (Ki67+ cells) as well as proliferating SCs (Ki67+/Pax7+ cells). The number of Ki67+ cells per biopsy was expressed relative to mm2 muscle tissue.

Gene expression (mRNA)

Detailed description of the method for RNA extraction and real-time reverse transcription polymerase chain reaction (PCR) is available in Supplemental Digital Content File (Document, Supplemental Digital Content 1, Methods, Primers for real-time reverse transcription PCR are listed in the Supplemental Table S2 (Supplemental Digital Content 3, primers for real-time PCR,


Anthropometric data, that is, changes in whole-body fat mass and whole-body lean mass, were analyzed with a two-tailed t-test for pairwise comparison. Functional test scores (4-m GST, DEMMI score, and 30-s CST) and the number of proliferating cells in muscle tissue sections were analyzed with the nonparametric Wilcoxon signed rank test. Unilateral muscle function (lower limb extension power and maximal knee extensor strength), leg lean mass, midthigh region lean mass, muscle thickness, muscle fiber CSA, and SC content were analyzed with a mixed two-way ANOVA (time (Pre vs Post, repeated-measures factor), treatment (E-Stim vs CON, group factor)), and the Holm–Sidak method was used for post hoc test of significant time–treatment interactions. In addition, data from scans of muscle mass were further analyzed for the difference between the individual relative changes in the E-Stim and CON leg with a two-tailed t-test for pairwise comparison. mRNA data were log transformed and analyzed with a two-tailed t-test for pairwise comparison of changes from Pre to Post within each leg as well as for the difference in changes between the legs. Data are presented as means ± SD, unless otherwise stated. A P value <0.05 was considered statistically significant. All statistical analyses were performed with SigmaPlot vs 13.0 (Systat Software Inc., San Jose, CA).


Descriptive data

Anthropometric data, functional test scores, and activity level during the hospital stay are presented in Table 1. Six of the patients (four men, two women) had an SMI below the cutoff level for sarcopenia.

Functional tests, muscle power, and maximal quadriceps strength

Four of the patients were not able to rise from a chair without using their arms (30-s CST = 0 repetitions). Three patients did not use a walking aid, one patient walked with a cane, seven patients used a rollator walker, and two patients used a high walking table. The average muscle power at Pre was 84 ± 51 W in the men (1.2 ± 0.5 W·kg−1) and 59 ± 36 W in the women (1.0 ± 0.5 W·kg−1). For comparison, leg extension power in 75- to 79-yr-old healthy individuals is considerably higher, that is, 148 ± 37 W (2.3 W·kg−1) in healthy elderly men and 80 ± 27 W (1.5 W·kg−1) in healthy elderly women (26). There were no significant changes in the whole-body functional tests during the hospitalization (DEMMI score, 30-s CST, 4-m GST, P = 0.32–0.78; Table 2). In the unilateral tests of muscle power and maximal knee extensor torque, there were no significant effect of time (P = 0.17–0.66), treatment (P = 0.25–0.68), or time–treatment interaction (P = 0.10–0.77; Table 2) in the E-Stim and CON leg.

Functional test scores, muscle function, lean mass, and muscle thickness.

DEXA scans and US scans

The mixed two-way ANOVA analysis revealed a significant time–treatment interaction for leg lean mass (P < 0.05), with a significant lower leg lean mass in the CON leg compared with the E-Stim leg after the intervention (−0.18 ± 0.24 kg, P < 0.05; Table 2). In addition, midthigh region lean mass showed a strong trend for a time–treatment interaction (P = 0.06; Table 2). An explorative post hoc test revealed a significantly lower midthigh region lean mass in the CON leg versus the E-Stim leg at Post (P < 0.05). There was no significant time–treatment interaction (P = 0.22) for VL + VI thickness measured by US (Table 2). The individual relative changes in leg lean mass from Pre to Post were significantly different between the E-Stim leg (−0.5% ± 5.0%) and the CON leg (−2.8% ± 5.3%, P < 0.05; Fig. 1). In addition, the individual relative changes in midthigh region lean mass were significantly different between the E-Stim (−0.7% ± 4.6%) and the CON leg (−2.4% ± 5.0%, P < 0.05; Fig. 1). The individual relative changes in VL + VI thickness were not significantly different between the E-Stim leg and the CON leg (P = 0.33; Fig. 1). There was no significant change in whole-body lean mass (−0.34 ± 1.49 kg, P = 0.43; Table 2), whereas whole-body fat mass was reduced from Pre to Post (−0.35 ± 0.45 kg, P < 0.05; Table 2).

Relative changes in leg lean mass and midthigh region lean mass measured with DEXA and the relative change in thickness of the thigh muscles (VL, VI) measured with US scan. Bars illustrate the mean of the individual relative changes in the CON leg (gray bars) and the E-Stim leg (white bars). The individual relative changes are illustrated with white symbols (men) and filled symbols (women) for the CON leg (circles) and the E-Stim leg (triangles) in n = 13 geriatric patients. *P < 0.05 between the legs.

Muscle fiber size, SCs, and proliferating cells

At the cellular level, there were no significant changes in muscle fiber size with hospitalization and E-Stim, as there were no main effects of time or treatment (P = 0.21–0.95) and no time–treatment interaction for type I fiber CSA (P = 0.54; Fig. 2A) or type II fiber CSA (P = 0.62; Fig. 2B). In addition, SC content was not affected by hospitalization and E-Stim, with no main effects of time or treatment (P = 0.55–0.90) and no time–treatment interaction in type I fibers (P = 0.44; Fig. 3E) or type II fibers (P = 0.72; Fig. 3F). The patient who had very few type II fibers in the Post biopsies (see Methods) is indicated in the figure (dotted lines), and removing this patient from the analysis did not change the outcome of the statistical tests. In the E-Stim leg, there was a significant increase in the number of Ki67+ cells per square millimeter muscle tissue from Pre to Post (P < 0.05; Fig. 3K), whereas no changes were found in the prevalence of Ki67+ cells per square millimeter muscle tissue in the CON leg (P = 0.13; Fig. 3K). Overall, there were few Ki67+/Pax7+ cells in the biopsies, with five Ki67+/Pax7+ cells in the total 390 Ki67+ cells identified. There was one Ki67+/Pax7+ cell in one Post E-Stim biopsy, and two Ki67+/Pax7+ cells in two Pre biopsies from the CON leg.

A and B, Muscle fiber size in type I fibers (A) and type II fibers (B), before (Pre) and after (Post) hospitalization in one leg receiving no treatment (CON) and the other leg receiving E-Stim. Bars illustrate mean fiber size for n = 9 patients in the CON leg (gray bars) and the E-Stim leg (white bars) at Pre and Post. Individual data at Pre and Post are illustrated with white symbols (men) and filled symbols (women) for the CON leg (circles) and the E-Stim leg (triangles). The dotted lines indicate the patient with very few type II fibers in two of the biopsies. C–E, Confocal images for muscle fiber CSA analysis are shown. The colors are pseudocolors, showing laminin staining of basement membrane (white) and myosin heavy chain I (MHC-I) staining of type I fibers (red). The vertical and horizontal yellow lines mark the overlap between neighboring tiles. Fibers included for analysis of fiber size are visualized in panels D and E as fibers delineated with a yellow line. Scale bars, 100 μm.
A–D, Wide-field microscopy images for SC analysis. The images show nuclei (DAPI, blue, A) and SCs (pax7, green, B), together with basement membrane (laminin, red, C) and MHC-I (cyan, D). The white arrow (A–D) indicates a cell considered to be an SC belonging to a type II fiber. The colors of laminin and MHC-I are pseudocolors. E–F, Number of SCs (Pax7+ cells) per fiber in type I fibers (E) and type II fibers (F), before (Pre) and after (Post) hospitalization in one leg receiving no treatment (CON) and the other leg receiving E-Stim. Bars illustrate mean number of SCs per fiber for n = 9 patients in the CON leg (gray bars) and the E-Stim leg (white bars) at Pre and Post. Individual data at Pre and Post are illustrated with white symbols (men) and filled symbols (women) for the CON leg (circles) and the E-Stim leg (triangles). F, The dotted lines indicate the patient with very few type II fibers in two of the biopsies. G–J, Wide-field microscopy images of staining for proliferating cells and SCs. The images show nuclei (DAPI, blue, G), proliferating cells (Ki67, red, H), SCs (pax7, green, I), and all channels combined (J). The white arrow indicates a Ki67-positive cell that is not pax7 positive (G, H, J). The yellow arrow indicates a pax7-positive cell that is not Ki67 positive (G, I, J). K, Number of Ki67-positive cells (Ki67+) per muscle tissue cross-section (mm2) in men (white symbols) and women (filled symbols) for the CON leg (circles) and the E-Stim leg (triangles). *P < 0.05 vs Pre. Scale bars, 100 μm.


Hospitalization did not affect gene expression in most of the genes measured in the CON leg except for a significant downregulation in MAFbx (P < 0.05) and a trend for a downregulation in MURF1 (P = 0.088; Fig. 4A). In the E-Stim leg, there was also a significant downregulation in MAFbx (P < 0.05), with a trend for a larger downregulation compared with the CON leg (P = 0.099), and MURF1 (P < 0.05) was significantly downregulated (Fig. 4A). There was furthermore a significant difference between the E-Stim and CON leg in the relative changes in myostatin gene expression (P < 0.05; Fig. 4A). Gene expressions of Collagen 1 (P < 0.001), TenascinC (P < 0.001), CD68 (P < 0.01), and Ki67 (P < 0.05) were significantly upregulated in the E-Stim leg only; there was a trend for a downregulation of HSP70 (P = 0.07; Fig. 4B); and there was a significant difference between the E-Stim and the CON leg (P < 0.001–0.05) for all these genes, together with a trend for a difference between the E-Stim and the CON leg in p16 (P = 0.095; Fig. 4C). No significant changes were observed in gene expression of GAPDH, IGF-1Ea, IGF-1Ec, Myogenin, NCAM, TCF7L2, and TNF-α (Fig. 4C + D).

Relative changes in gene expression levels (mRNA) in muscle biopsies in geriatric patients (n = 9) collected in the same leg at admission and discharge from the hospital in the leg receiving no special treatment (CON; gray bars) and the leg receiving E-Stim (white bars). RPLP0 mRNA was chosen as internal control (housekeeping gene), and GAPDH mRNA was measured and normalized with RPLP0 to validate the use of RPLP0, with no difference for GAPDH, supporting the use of RPLP0 for normalization. Symbols in brackets () denotes a trend for significance at P = 0.05–0.10. *P < 0.05 from Pre within the same leg; #P < 0.05 between the relative changes in the E-Stim and CON leg. Data are geometric means ± back-transformed SEM displayed on a logarithmic scale y axis (Log2).


The present study is to our knowledge the first study to investigate the effects of hospitalization upon skeletal muscle either with or without daily sessions of E-Stim–induced muscle activation in patients admitted to a geriatric ward. DEXA scans revealed a significant positive effect of E-Stim in preserving leg lean mass and midthigh region lean mass. Functional test scores and muscle function did not change during the study. Although E-Stim increased the prevalence of proliferating cells in muscle biopsies from a subgroup of nine patients, there were no changes in muscle fiber size, the size of the SC pool, and the prevalence of activated SCs. However, compared with the CON leg, E-Stim resulted in overall favorable changes in gene expression associated with muscle fiber atrophy (myostatin, MURF1, MAFbx), extracellular matrix remodeling (Collagen 1, TenascinC), and markers of cell activity (CD68, Ki67), together demonstrating a positive response at the cellular level.

Muscle function and physical function

When indirectly assessed with interviews, 35%–65% of elderly medical patients have experienced functional decline before admission to the hospital (27,28), and more than 50% of the patients are discharged with a decline in their functional level compared with their preadmission level (27,28). However, several studies do not observe functional decline during hospitalization in various measures of muscle function and mobility (16,29,30), possibly reflecting recovery from the patient’s medical condition. In line with these observations, we did not observe any decline in measures of muscle strength and power in the CON leg. In addition, there was no decline in whole-body functional test scores, but it should be noted that these tests were included for general descriptive purposes, whereas they are not suited for evaluation of a unilateral intervention as they include bilateral leg muscle activation. Although longer duration E-Stim interventions can improve functional performance and muscle strength (31), there was no measurable effect of the E-Stim on these parameters in the present study. This was somewhat expected because of the short duration of the intervention in the present study and in line with our previous results after heavy resistance training with a comparable study design (5).

Lean mass

Although the number of days spent in a hospital during 1 yr is related to a decline in both lean mass and strength in elderly individuals (2), it is not well explored whether this actually occurs during the hospital stay. Two studies in acutely admitted geriatric patients have reported no decline in either leg lean mass (5) or whole-body potassium (6). In the present study, the relative decline in leg lean mass and midthigh region lean mass was significantly larger in the CON leg compared with the E-Stim leg (Fig. 1), and the CON leg had significantly lower leg lean mass versus the E-Stim leg at the end of the hospital stay (Table 2). These results suggest a modest loss of muscle mass in the lower extremities in these patients. Although many factors may contribute to hospital associated loss of muscle mass, that is, activity level, nutritional status, and systemic factors related to the medical illness (3,4,11), one daily session of E-Stim–induced muscle contractions efficiently preserved lean mass in the present study. This is a very important observation, in line with our previous report of a positive effect of resistance training on lean mass in patients admitted to the same geriatric ward (5). Together, these data demonstrate that geriatric patients respond to muscle activation during hospitalization in a similar positive manner to healthy elderly individuals during a period of inactivity (14,15).

Noteworthy, the ~2.4% greater relative decline in leg lean mass in the CON leg corresponds to a daily decline in leg lean mass of approximately 0.4% in the absence of E-Stim. Disuse atrophy models in healthy elderly individuals result in a 0.3%–0.7% decline per day in quadriceps volume with lower limb immobilization (13,32), a 0.6%–0.8% decline per day in leg lean mass with bed rest (12,15,33,34), and a ~0.3% decline per day in leg lean mass in a model of reduced activity where step count was reduced to 1500 steps per day (35). The ~0.4% decline per day in leg lean mass in the absence of E-Stim in the present study therefore seems to be within the range of what is observed in disuse atrophy models in healthy elderly individuals. However, it is important to emphasize that the present results may underestimate the average loss of muscle mass in these patients for several reasons. First of all, the present study design included a demanding test battery, and the included patients may have been a selected group of patients, as they were younger, had better functional test scores, and had slightly higher activity level, compared with a larger group of patients included in an observational study the year before at the same department (16). For ethical and practical reasons, the patients were pretested on their second day at the geriatric department, many of the patients had already spent ~1 d at the acute care ward plus they may have been sick and inactive for 1–2 d before arriving at the hospital, and we may therefore have missed the time point where the rate of loss of muscle mass peaked. Furthermore, because the posttest was performed 6.6 ± 1.0 d after the pretest, the duration of the intervention was considerably shorter than the average length of stay at this department (~10 d) at the time of the study (30). Finally, it should be noted that changes in lean mass (DEXA) can also be affected simply by a change in hydration status, for example, in patients who were dehydrated at the time of admission to the hospital and became rehydrated during the hospital stay. However, the unilateral within-subject design accounts for this regarding differences between the legs, so these potential limitations do not negate our findings of a preserved lean mass with E-Stim in geriatric patients during a hospital stay.

At the cellular level

At the cellular level, we hypothesized that E-Stim could preserve muscle fiber size, whereas we did not expect any significant muscle fiber hypertrophy within the short duration of this intervention. Because of the lack of a significant decline in muscle fiber size in the CON leg, this hypothesis could not be confirmed. The lack of significant muscle fiber atrophy contrasted with the ~2.5% decline in leg lean mass in the DEXA scans. Significant muscle fiber atrophy together with a ~4% decline in thigh lean mass has previously been reported in healthy elderly individuals after short-term bed rest (15), whereas others have failed to detect significant muscle fiber atrophy despite a ~3.5% decline in quadriceps CSA after 5 d of immobilization (19). The present data therefore probably reflect that changes in muscle fiber size were smaller than what could be detected with the muscle biopsy technique in these patients. Interestingly, the changes from Pre to Post in the expression of genes related to the proteolytic ubiquitin signaling pathway overall indicated a further downregulation of MURF1 and MAFbx in the E-Stim leg compared with the CON leg. In addition, E-Stim induced favorable changes in gene expression of myostatin, a member of the TGF-β family and a negative regulator of muscle growth, with a reduced expression in the E-Stim leg versus the CON leg. Somewhat surprising, although bed rest and immobilization are often associated with elevated gene expression of MURF1, MAFbx, and/or myostatin in healthy elderly individuals (7,15,34,36), we observed a downregulation of MURF1 and MAFbx in the CON leg during the hospital stay. Elevated mRNA levels of MURF1 and MAFbx have previously been reported in critically ill patients compared with healthy controls (18,37), and although the patients in the present study may not be comparable to critically ill patients, the present data probably reflect that MURF1 and MAFbx mRNA levels were elevated in the geriatric patients at the time of admission to the department. The relative decline in gene expression from admission to discharge may therefore reflect that the patients had recovered from their illness during the hospital stay. Without a healthy age and sex-matched control group, we are unfortunately not able to determine if MURF1 and MAFbx mRNA levels were elevated at the time of the Pre biopsy. In addition, it is not clear from these data whether MURF1 and MAFbx mRNA levels had returned to a normal level during the hospital stay. It could be speculated that the changes were the result of an increased activity level during the hospitalization, but day-to-day activity recordings showed no difference from the first to the last day of the hospital stay. However, it cannot be ruled out that with the medical treatment and daily support from the care personnel at the geriatric ward, the patients were able to be more physically active at the geriatric department compared with the acute care ward and the days leading up to the hospital admission. E-Stim resulted in a further and possibly earlier downregulation of these genes, which may be an important mechanism for the preservation of muscle mass. In contrast, E-Stim only resulted in small and nonsignificant increases in gene expression of IGF-1Ea (P = 0.203) and IGF-1Ec (P = 0.127), members of the IGF-1 family that are believed to play an important role in muscle hypertrophy, SC activation, proliferation, and differentiation (38), and also exert an inhibiting effect on the ubiquitin pathway (39). Upregulation in these IGF-1 isoforms was reported 7 d after the last training session of a 9-wk E-Stim protocol in healthy elderly individuals (31). In the present study, we may have missed the time point for a peak increase in IGF-1 expression, but it is also possible that the intervention was too short to induce significant changes.

Immunohistochemical analysis of SC content revealed that SCs were not lost in the CON leg during hospitalization, and that E-Stim did not increase the SC content. It is important to emphasize that it is not fully clear whether SCs are lost during disuse atrophy (9,10). Without a significant reduction in the SC pool in the CON leg in the present study, it is not possible to conclude whether E-Stim could have preserved the SC pool in these patients. E-Stim could in addition be a potent stimulus for SC activation and proliferation, potentially elevating the SC content of the stimulated muscles, which may be beneficial for subsequent rehabilitation (7). However, although gene expression of the cell proliferation marker Ki67+ increased with E-Stim, and there was a significant increase in the prevalence of Ki67+ cells per mm2 muscle, there were no increase in SC content and activated SCs (Ki67+/Pax7+ cells) in the E-Stim leg and no increase in gene expression of myogenin and NCAM, suggesting a poor SC response to this intervention, at least at the time point of the Post biopsies. Our laboratory has previously used the Ki67 antibody to label activated SCs (Pax7+/Ki67+) cells after a severe muscle damaging protocol that resulted in a substantial increase in the number of SCs per fiber but only a modest increase in the proportion of Ki67+/Pax7+ cells 7 d after the intervention (40). Noteworthy, the increase in Ki67+/Pax7+ cells was not significant 48 h after this muscle damaging protocol (40). In the present study, as part of the immunohistochemical staining, the Ki67 antibody was also tested on a sample with many Pax7+/Ki67+ cells from that study, confirming that the Ki67 antibody worked. It should be noted that the E-Stim intensity had to be increased gradually from session to session, as the tolerance to the stimulation gradually improved in the patients. The biopsies were taken in the deep portion of VL, and it is likely that it took some days before the E-Stim intensity had reached a level high enough to recruit muscle fibers in this area. However, the mRNA data strongly suggest that muscle fibers in the biopsied region had been recruited during the stimulation. The lack of SC activation may furthermore be related to a blunted or delayed SC responsiveness with aging (41). Therefore, although speculative, the present data cannot rule out the possibility that the E-Stim protocol could have resulted in activated SCs with a more prolonged intervention. The increase in Ki67 mRNA levels as well as Ki67+/Pax7 cells per mm2 demonstrates that other cell types, for example, immune cells, endothelial cells, or fibroblasts, were activated in response to the E-Stim. Based on the increase in CD68 mRNA, we speculate that the activated cells were primarily macrophages, whereas the unchanged expression of TCF7L2 may indicate that the activated cells were not fibroblasts. However, it should be noted that although CD68 is frequently used as a marker of macrophages, it is also expressed in fibroblasts and endothelial cells (42), but it was not within the aim of the present study to quantify these cell types in detail. Overall, the gene expression data therefore seem to reflect an acute stress response and an increase in proliferating macrophages, further supported by the increased expression of Collagen I and TenascinC, genes related to extracellular matrix remodeling that are elevated after muscle damaging exercise (43). It should be noted that we did not register muscle soreness in the patients at any time during the study and found no muscle fibers with a high density of central nuclei, a sign of myofiber necrosis and macrophage infiltration (44). In addition, there was a trend for a downregulation of the expression of HSP70, a marker of cell stress that is upregulated as an early response to myofiber damage (43). The elevated gene expression of Collagen I and TenascinC may instead reflect a general response to the mechanical stress during forceful muscle contractions and a gradual adaptation of the myofibers and extracellular matrix to this type of stimulation.


The present study demonstrates the positive effects of E-Stim–induced muscle contractions, preserving lean mass in geriatric patients during hospitalization and resulting in favorable changes at the gene expression level with downregulation of genes associated with muscle protein breakdown. E-Stim may be the only option for muscle activation in patients who are not easily mobilized for various reasons, but similar effects may be achieved with structured functional training programs or with heavy resistance training in patients where this is feasible. Altogether the changes in gene expression in the present study demonstrate that E-Stim induced downregulation of genes involved in muscle atrophy, which may be related to the preservation of leg lean mass in the E-Stim leg.

This work was supported by The Nordea Foundation (Healthy Aging Grant), The Danish Council for Independent Research, Bispebjerg Hospital, The A.P. Møller Foundation for the Advancement of Medical Science, and The Faculty of Health and Medical Sciences at the University of Copenhagen.

The staff at the Geriatric Department are gratefully acknowledged for their contribution during the study. We thank Camilla Sørensen and Anja Jokipi-Nielsen at the Institute of Sports Medicine Copenhagen for technical assistance in preparing the muscle biopsy samples, and Rasmus L. Bechshoeft and Christian S. Eriksen for assistance in collecting the biopsies. René B. Svensson at the Institute of Sports Medicine Copenhagen is warmly acknowledged for his development of the ImageJ macro, and we acknowledge the Core Facility for Integrated Microscopy, Faculty of Health and Medical Sciences, University of Copenhagen, where the confocal images were taken.

No conflicts of interest, financial or otherwise, are declared by the authors. The results of the present study do not constitute endorsement by the American College of Sports Medicine. All authors declare that the results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.


1. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al. Sarcopenia: European consensus on definition and diagnosis: report of the European Working Group on Sarcopenia in Older People. Age Ageing. 2010;39(4):412–23.
2. Alley DE, Koster A, Mackey D, et al. Hospitalization and change in body composition and strength in a population-based cohort of older persons. J Am Geriatr Soc. 2010;58(11):2085–91.
3. English KL, Paddon-Jones D. Protecting muscle mass and function in older adults during bed rest. Curr Opin Clin Nutr Metab Care. 2010;13(1):34–9.
4. Welch C, K Hassan-Smith Z, A Greig C, M Lord J, A Jackson T. Acute sarcopenia secondary to hospitalisation—an emerging condition affecting older adults. Aging Dis. 2018;9(1):151–64.
5. Norheim KL, Cullum CK, Andersen JL, Kjaer M, Karlsen A. Inflammation relates to resistance training–induced hypertrophy in elderly patients. Med Sci Sports Exerc. 2017;49(6):1079–85.
6. Bautmans I, Njemini R, Lambert M, Demanet C, Mets T. Circulating acute phase mediators and skeletal muscle performance in hospitalized geriatric patients. J Gerontol A Biol Sci Med Sci. 2005;60(3):361–7.
7. Suetta C, Frandsen U, Mackey AL, et al. Ageing is associated with diminished muscle re-growth and myogenic precursor cell expansion early after immobility-induced atrophy in human skeletal muscle. J Physiol. 2013;591(15):3789–804.
8. Snijders T, Nederveen JP, McKay BR, et al. Satellite cells in human skeletal muscle plasticity. Front Physiol. 2015;6:283.
9. Arentson-Lantz EJ, Paddon-Jones D, Fry CS. The intersection of disuse-induced muscle atrophy and satellite cell content: reply to Snijders, Nederveen, and Parise. J Appl Physiol (1985). 2016;120(12):1491.
10. Snijders T, Nederveen JP, Parise G. Are satellite cells lost during short-term disuse-induced muscle fiber atrophy? J Appl Physiol (1985). 2016;120(12):1490.
11. Evans WJ. Skeletal muscle loss: cachexia, sarcopenia, and inactivity. Am J Clin Nutr. 2010;91(4):1123S–7.
12. Kortebein P, Ferrando A, Lombeida J, Wolfe R, Evans WJ. Effect of 10 days of bed rest on skeletal muscle in healthy older adults. JAMA. 2007;297(16):1772–4.
13. Suetta C, Hvid LG, Justesen L, et al. Effects of aging on human skeletal muscle after immobilization and retraining. J Appl Physiol (1985). 2009;107(4):1172–80.
14. Devries MC, Breen L, Allmen MV, et al. Low-load resistance training during step-reduction attenuates declines in muscle mass and strength and enhances anabolic sensitivity in older men. Physiol Rep. 2015;3(8):e12493.
15. Reidy PT, McKenzie AI, Brunker P, et al. Neuromuscular electrical stimulation combined with protein ingestion preserves thigh muscle mass but not muscle function in healthy older adults during 5 days of bed rest. Rejuvenation Res. 2017;20(6):449–61.
16. Karlsen A, Loeb MR, Andersen KB, et al. Improved functional performance in geriatric patients during hospital stay. Am J Phys Med Rehabil. 2017;96(5):e78–84.
17. Magnus CRA, Barss TS, Lanovaz JL, Farthing JP. Effects of cross-education on the muscle after a period of unilateral limb immobilization using a shoulder sling and swathe. J Appl Physiol (1985). 2010;109(6):1887–94.
18. Dirks ML, Hansen D, Van Assche A, Dendale P, Van Loon LJC. Neuromuscular electrical stimulation prevents muscle wasting in critically ill comatose patients. Clin Sci (Lond). 2015;128(6):357–65.
19. Dirks ML, Wall BT, Snijders T, Ottenbros CLP, Verdijk LB, van Loon LJC. Neuromuscular electrical stimulation prevents muscle disuse atrophy during leg immobilization in humans. Acta Physiol (Oxf). 2014;210(3):628–41.
20. de Morton NA, Davidson M, Keating JL. The DeMorton Mobility Index (DEMMI): an essential health index for an ageing world. Health Qual Life Outcomes. 2008;6:63.
21. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85–94.
22. Jones CJ, Rikli RE, Beam WC. A 30-s chair-stand test as a measure of lower body strength in community-residing older adults. Res Q Exerc Sport. 1999;70(2):113–9.
23. Akima H, Kubo K, Kanehisa H, Suzuki Y, Gunji A, Fukunaga T. Leg-press resistance training during 20 days of 6° head-down-tilt bed rest prevents muscle deconditioning. Eur J Appl Physiol. 2000;82(1–2):30–8.
24. Vahlgren J, Karlsen A, Scheel FU, et al. Using ultrasonography to detect loss of muscle mass in the hospitalized geriatric population. Transl Sports Med. 2019;2(5):287–93.
25. Karlsen A, Bechshøft RL, Malmgaard-Clausen NM, et al. Lack of muscle fibre hypertrophy, myonuclear addition, and satellite cell pool expansion with resistance training in 83–94-year-old men and women. Acta Physiol (Oxf). 2019;227(1): e13271.
26. Skelton DA, Greig CA, Davies JM, Young A. Strength, power and related functional ability of healthy people aged 65–89 years. Age Ageing. 1994;23(5):371–7.
27. Covinsky KE, Palmer RM, Fortinsky RH, et al. Loss of independence in activities of daily living in older adults hospitalized with medical illnesses: increased vulnerability with age. J Am Geriatr Soc. 2003;51(4):451–8.
28. Mudge AM, O’Rourke P, Denaro CP. Timing and risk factors for functional changes associated with medical hospitalization in older patients. J Gerontol A Biol Sci Med Sci. 2010;65(8):866–72.
29. Bodilsen AC, Pedersen MM, Petersen J, et al. Acute hospitalization of the older patient: changes in muscle strength and functional performance during hospitalization and 30 days after discharge. Am J Phys Med Rehabil. 2013;92(9):789–96.
30. Oestergaard AS, Mathiesen MH, Karlsen A, et al. In acutely admitted geriatric patients, offering increased physical activity during hospitalization decreases length of stay and can improve mobility. Transl Sports Med. 2018;1(1):46–53.
31. Kern H, Barberi L, Löfler S, et al. Electrical stimulation counteracts muscle decline in seniors. Front Aging Neurosci. 2014;6:189.
32. Dideriksen K, Boesen AP, Kristiansen JF, et al. Skeletal muscle adaptation to immobilization and subsequent retraining in elderly men: no effect of anti-inflammatory medication. Exp Gerontol. 2016;82:8–18.
33. Drummond MJ, Dickinson JM, Fry CS, et al. Bed rest impairs skeletal muscle amino acid transporter expression, mTORC1 signaling, and protein synthesis in response to essential amino acids in older adults. Am J Physiol Endocrinol Metab. 2012;302(9):E1113–22.
34. Tanner RE, Brunker LB, Agergaard J, et al. Age-related differences in lean mass, protein synthesis and skeletal muscle markers of proteolysis after bed rest and exercise rehabilitation. J Physiol. 2015;593(18):4259–73.
35. Breen L, Stokes KA, Churchward-Venne TA, et al. Two weeks of reduced activity decreases leg lean mass and induces “anabolic resistance” of myofibrillar protein synthesis in healthy elderly. J Clin Endocrinol Metab. 2013;98(6):2604–12.
36. Suetta C, Frandsen U, Jensen L, et al. Aging affects the transcriptional regulation of human skeletal muscle disuse atrophy. PLoS One. 2012;7(12):e51238.
37. Constantin D, McCullough J, Mahajan RP, Greenhaff PL. Novel events in the molecular regulation of muscle mass in critically ill patients. J Physiol. 2011;589(Pt 15):3883–95.
38. Kraemer WJ, Ratamess NA, Nindl BC. Recovery responses of testosterone, growth hormone, and IGF-1 after resistance exercise. J Appl Physiol. 2016;122(3):549–58.
39. Sandri M, Sandri C, Gilbert A, et al. Foxo transcription factors induce the atrophy-related ubiquitin ligase atrogin-1 and cause skeletal muscle atrophy. Cell. 2004;117(3):399–412.
40. Mackey AL, Rasmussen LK, Kadi F, et al. Activation of satellite cells and the regeneration of human skeletal muscle are expedited by ingestion of nonsteroidal anti-inflammatory medication. FASEB J. 2016;30(6):2266–81.
41. Snijders T, Verdijk LB, Smeets JSJ, et al. The skeletal muscle satellite cell response to a single bout of resistance-type exercise is delayed with aging in men. Age (Dordr). 2014;36(4):9699.
42. Gottfried E, Kunz-Schughart LA, Weber A, et al. Expression of CD68 in non-myeloid cell types. Scand J Immunol. 2008;67(5):453–63.
43. Mackey AL, Kjaer M. Connective tissue regeneration in skeletal muscle after eccentric contraction-induced injury. J Appl Physiol (1985). 2017;122(3):533–40.
44. Mackey AL, Kjaer M. The breaking and making of healthy adult human skeletal muscle in vivo. Skelet Muscle. 2017;7(1):24.


Supplemental Digital Content

© 2020 American College of Sports Medicine