I'm stuck in bed for two weeks — how much leg muscle will I lose?
In 12 healthy young men kept in strict bed rest for 14 days, thigh muscle volume fell 5% on MRI (7.1 → 6.7 L), leg lean mass 5% on DXA (10.2 → 9.7 kg) and thigh cross-sectional area 6% on CT, with the quadriceps side losing 7%. That is strict bed rest — wheelchair to the scanner — not a fortnight of normal life without training, and no strength or performance was measured.
Fuchs et al. re-used the scans of 12 healthy young men who spent 14 days in strict bed rest (never out of bed; wheelchair to the scanners) in a trial of daily one-leg blood-flow restriction that had shown no effect, and averaged both legs.
Leg lean mass on DXA fell 5% (10.2 → 9.7 kg, −546 g), thigh muscle cross-sectional area on single-slice CT fell 6% (155 → 146 cm²), and thigh muscle volume on MRI fell 5% whether segmented by hand (7.1 → 6.7 L) or automatically (7.2 → 6.8 L). The anterior compartment (quadriceps, sartorius, tensor fasciae latae) lost 7% by manual MRI, the posterior 4%.
Automated MRI agreed with manual analysis (r = 0.96, bias −0.11 ± 0.29 L, ICC 0.96), so the authors call manual analysis obsolete for whole-thigh atrophy. No strength or performance was measured.
Protocol
Participants: 12 healthy men, 24 ± 3 y, 79.4 ± 12.4 kg, 1.83 ± 0.07 m, BMI 23.7 ± 3.1 kg/m²; from a randomised trial (registered NL6222) of daily unilateral blood-flow restriction during bed rest, which had no effect on any leg measure.
Bed rest: 14 days strict — no leaving the bed for any reason, hygiene in bed; standardised, controlled diet; transported by wheelchair to the post-bed-rest scans.
Scans (before and after, fasted, bladder and bowels emptied): DXA leg lean mass (CV 1.3%); single-slice CT 15 cm above the patella (thigh muscle CSA, blinded, CV 0.6%); 3T MRI Dixon neck-to-knee, thigh muscle volume (all thigh and some pelvic muscles) and anterior and posterior compartments, segmented manually (ITK-SNAP, about five working days per person per time point) and automatically (AMRA Researcher), blinded and independent.
Analysis: left and right legs averaged for atrophy (12 data points); legs treated separately for method comparison (24 data points); paired t-tests, Pearson r, Bland-Altman, ICC.
Key Results
| Measure | Before | After | Change | p |
|---|---|---|---|---|
| DXA leg lean mass | 10.2 ± 1.6 kg | 9.7 ± 1.6 kg | −5% (−546 ± 165 g) | < 0.001 |
| CT thigh muscle CSA | 155 ± 26 cm² | 146 ± 24 cm² | −6% (−9.2 ± 4.1 cm²) | < 0.001 |
| MRI thigh muscle volume, manual | 7.1 ± 1.1 L | 6.7 ± 1.0 L | −5% (−372 ± 214 mL) | < 0.001 |
| MRI thigh muscle volume, automated | 7.2 ± 1.1 L | 6.8 ± 1.0 L | −5% (−360 ± 108 mL) | < 0.001 |
| MRI anterior thigh, manual | 2.8 ± 0.4 L | 2.6 ± 0.4 L | −7% (−183 ± 112 mL) | < 0.001 |
| MRI anterior thigh, automated | 2.7 ± 0.4 L | 2.6 ± 0.4 L | −6% (−177 ± 58 mL) | < 0.001 |
| MRI posterior thigh, manual | 4.3 ± 0.7 L | 4.1 ± 0.7 L | −4% (−185 ± 125 mL) | < 0.001 |
| MRI posterior thigh, automated | 4.4 ± 0.7 L | 4.2 ± 0.7 L | −4% (−183 ± 71 mL) | < 0.001 |
Manual vs automated MRI, total thigh: r = 0.96 (95% CI 0.93 to 0.98), bias −0.11 ± 0.29 L, limits of agreement −0.67 to 0.45 L, ICC 0.96 (0.92-0.98), CV 2.74%; atrophy did not differ between methods (p = 0.920).
MRI vs DXA r = 0.95 (manual) and 0.98 (automated); MRI vs CT r = 0.74 and 0.79.
Mechanisms
Measured: muscle loss was larger in the anterior compartment, dominated by the quadriceps (−7%), than in the posterior (−4%).
Authors' interpretation: the quadriceps is more susceptible to disuse atrophy than other thigh muscles, in line with earlier bed-rest work; lower-body atrophy is what compromises walking and mobility.
Not measured: muscle protein synthesis, strength, neural drive or performance; the mechanism of the loss is not examined in this paper.
Inference: two weeks of total disuse took roughly what 12 weeks of novice training added in DXA leg lean mass (+5.0%) and about half of the quadriceps gain (+13.1% CSA) in REF-1136 (different people and methods).
Practical Application
I'm stuck in bed for two weeks — how much leg muscle will I lose? In 12 healthy young men kept in strict bed rest for 14 days, thigh muscle volume fell 5% on MRI (7.1 → 6.7 L), leg lean mass 5% on DXA (10.2 → 9.7 kg) and thigh cross-sectional area 6% on CT, with the quadriceps side losing 7%. That is strict bed rest — wheelchair to the scanner — not a fortnight of normal life without training, and no strength or performance was measured.
Move whatever you are allowed to move. The loss here came with zero ambulation; the base's bed-rest entry lists the countermeasures tested so far, and only fluid and salt loading has been tried in trained people (REF-1134). What is allowed during an illness or after surgery is the treating doctor's call.
Plan the rebuild around the quadriceps. The front of the thigh lost most; a return that includes strength work, not only riding or walking, matches the 2-4-week rehabilitation with resistance exercise in REF-1134 (Tier 0-1 data).
A body-composition scan can see a loss this size. DXA, CT and MRI all detected the ~5% change; bathroom-scale impedance was not tested.
Limitations
Population: 12 healthy young men; training status not reported here; no women, no older adults, no athletes.
No control group, no function: an uncontrolled pre-post measurement; the paper measures muscle quantity only.
Strict bed rest: results apply to total immobility with controlled diet, not to illness, an injured limb in a cast, or simply not training.
One leg had daily blood-flow restriction in the parent trial; it had no effect, so both legs were averaged.
DXA lean mass is not pure muscle: it includes water and is sensitive to hydration and food, which the study controlled.
Conflicts: two authors are employees of AMRA Medical AB (the automated MRI software) and two collaborated with and received research support from it; funding for scanning from Stimuleringsfonds FHML, Maastricht.
How this entry was checked and changed
Why this level
uncontrolled pre-post, n = 12, one lab; the measurement comparison is careful (blinded, independent analyses) but there is no control group and no functional outcome; read from the full text
Full text (checked 2026-10-15)
"Twelve healthy young men (age: 24 ± 3 years; body mass: 79.4 ± 12.4 kg; height: 1.83 ± 0.07 m; BMI: 23.7 ± 3.1 kg/m2)" (full text).
"all participants were subjected to 2 weeks of strict bed rest and were not permitted to leave the bed for any reason"; "transported in a wheelchair" (full text).
"we did not observe any effect of blood flow restriction on any leg muscle assessment" (full text).
"DXA‐derived leg lean mass declined by 5% (546 ± 165 g) following bed rest, from 10.2 ± 1.6 to 9.7 ± 1.6 kg (p < 0.001" (full text).
"thigh muscle cross‐sectional area declined by 6% (9.2 ± 4.1 cm2) following bed rest, from 155 ± 26 to 146 ± 24 cm2" (full text).
Manual MRI thigh volume "declined by 5% (372 ± 214 mL) following bed rest, from 7.1 ± 1.1 to 6.7 ± 1.0 L"; automated "by 5% (360 ± 108 mL) following bed rest, from 7.2 ± 1.1 to 6.8 ± 1.0 L"; methods "(p = 0.920)" (full text).
Anterior manual "declined by 7% (183 ± 112 mL) following bed rest, from 2.8 ± 0.4 to 2.6 ± 0.4 L"; automated "decline of 6% (177 ± 58 mL)" (full text).
Posterior manual "declined by 4% (185 ± 125 mL) following bed rest, from 4.3 ± 0.7 to 4.1 ± 0.7 L"; automated "decline by 4% (183 ± 71 mL) ... from 4.4 ± 0.7 to 4.2 ± 0.7 L" (full text).
Agreement: "r = 0.96, 95% CI: 0.93 to 0.98"; "bias (−0.11 ± 0.29 L) and 95% limits of agreement being between −0.67 and 0.45 L"; "ICC ... was 0.96 (95% CI: 0.92–0.98), with a CV of 2.74" (full text).
Table 1: manual MRI vs DXA "0.95"; automated vs DXA "0.98"; manual vs CT "0.74"; automated vs CT "0.79" (full text).
DXA CV "1.3%"; CT CV "0.6%"; CT slice "15 cm proximal to the top of the patella" (full text).
Manual segmentation "required an average of five full working days per individual for each time point" (full text).
"Per Widholm and Olof Dahlqvist Leinhard are employees of AMRA Medical AB" (full text).