In 85 untrained older adults, one set of knee extension twice a week left 60% with no measurable quadriceps growth after 10 weeks, while four sets on the other leg grew those same people's muscle (+3.02 cm² against +0.49 cm², not significant) and doubled their strength gain (+6.11 kg against +2.98 kg); in those who grew on one set, four sets added a non-significant extra
Main caveat: One exercise (knee extension) and one muscle region (mid-thigh CSA); multi-joint lifts and the whole muscle length were not studied.
Lixandrão and colleagues asked whether more sets rescue people who do not grow on a low dose. Eighty-five untrained adults over 60 trained knee extension twice a week for 10 weeks, one leg with 1 set and the other with 4 sets, every set to concentric failure at 8-15 RM. Quadriceps cross-sectional area was measured by MRI and strength as one-leg 1RM.
People whose 1-set leg grew by no more than twice the MRI typical error (2 × TE = 3.27%) were called non-responders: 51 of 85 (60%). In them, the 4-set leg grew while the 1-set leg did not, and strength rose more on 4 sets. In the 34 responders both legs grew and strength rose about equally, with a tendency for more growth on 4 sets.
The authors conclude that higher volume can offset non-response to a low dose in older adults.
Participants
186 screened, 116 randomised, 102 allocated after 14 withdrew before baseline; 14 stopped for personal reasons and 3 were removed for having trained before; 85 analysed (41 men, 44 women), age 68 ± 4 y, BMI 26.4 ± 3.7.
Untrained: no regular resistance or aerobic training for at least 6 months. Excluded: type 1 diabetes, ischaemic heart disease, arrhythmia, uncontrolled hypertension, major orthopaedic disorders, anabolic supplements or drugs.
Protein: everyone took two 20-g whey doses a day; intake 1.6 ± 0.3 g/kg/day (non-responders) and 1.5 ± 0.4 (responders), no group difference. Adherence 97.3 ± 4.5% and 98.5 ± 2.6%.
Protocol
Unilateral knee extension, 2 sessions a week for 10 weeks: one leg 1 set (2 sets a week), the other 4 sets (8 a week), 8-15 RM, 60-90 s between sets, every set to concentric failure, supervised.
MRI quadriceps cross-sectional area at mid-thigh, before and after; one-leg 1RM after one familiarisation session.
Classification by the 1-set leg's change: non-responder ≤ 2 × TE (3.27%), responder > 2 × TE.
Key Results
| Group (n) | Quadriceps CSA change, 1 set | 4 sets | 1RM change, 1 set | 4 sets |
|---|---|---|---|---|
| Non-responders (51) | +0.49 cm² (95% CI −0.46 to 1.46; P = 0.54) | +3.02 cm² (2.06-3.99) | +2.98 kg (0.88-5.08) | +6.11 kg (4.01-8.21) |
| Responders (34) | +2.81 cm² (1.41-4.21) | +4.15 cm² (2.75-5.55); set × time P = 0.08 | +4.54 kg (1.79-7.29) | +6.29 kg (3.54-9.04); set × time P = 0.23 |
Non-responders: set × time interaction significant for both muscle size and strength (P < 0.05); after training the 4-set leg was 2.12 cm² larger than the 1-set leg (1.15-3.19).
Responders out-grew non-responders on both doses (1 set: 2.31 cm², 4 sets: 1.12 cm²).
The printed CI for the non-responders' 1-set change reads "0.46 to 1.46" around a mean of 0.49 with P = 0.54; the lower bound must be −0.46 (our reading: a lost minus sign).
Practical Application
For an older beginner, one set is not enough for most people's muscle. Six in ten grew no measurable muscle on one set to failure twice a week for 10 weeks; four sets grew it in the same people. Plan more than one set per exercise when muscle size is the goal at 60 and over (SYNTH-105).
Strength rose on one set in everyone, but more on four in those who did not grow. For strength alone, one set did something in all; four did more in the majority.
Inference, not tested here: a low responder on a small dose may be under-dosed rather than unable to respond; raise sets before concluding someone "does not respond".
Not from this paper: how many sets after the first 10 weeks, and whether four sets is the best number — only two doses were tested.
Limitations
One exercise (knee extension) and one muscle region (mid-thigh CSA); multi-joint lifts and the whole muscle length were not studied.
Classification used the 1-set leg's own noisy change; part of the "non-response" may be measurement error or regression to the mean, which the 4-set leg, measured independently, partly guards against.
Unilateral training can transfer some strength to the other leg (cross-education); the authors cite that it is small when both legs train.
Untrained adults over 60 with protein supplemented; not generalisable to trained or younger lifters.
Only two doses and 10 weeks.
How this entry was checked and changed
Why this level
every participant is their own control, which the rubric treats like a crossover (≥15 → B+, no route above B+); n = 85 analysed of 102 allocated; read in full; one exercise, one muscle region measured
Content correction 2026-11-06
2026-11-06: added from the full text (the editor's PDF), with the Roberts 2026 table row that already named this work; level B+ (within-subject randomised, n = 85).