REF-484 · Coyle et al., 1992 · Medicine & Science in Sports & Exercise, · Paper

Cycling efficiency is related to the percentage of type I muscle fibres

Blimited evidenceMeasured on highly trained athletesWritten before October 2026, being re-checked

Participants

n=19 highly endurance-trained male cyclists, VO₂max 4.9 ± 0.1 L/min, body weight 71 ± 1 kg (about 69 ml/kg/min by division)

Protocol

Vastus lateralis biopsy (ATPase histochemistry) for fibre-type composition; steady-state cycling on a Monark ergometer at 80 rpm at 52, 61 and 71% VO₂max. Gross efficiency (GE) = work ÷ caloric expenditure; delta efficiency (DE) = slope of that relationship between ~50 and 70% VO₂max. A sub-group (n=13) also did two-legged knee extension at 50 and 70% of task-peak VO₂.

Key Findings

% Type I fibres ranged 32-76%; delta efficiency ranged 18.3-25.6% — large between-athlete spread in efficiency even among highly trained cyclists.
% Type I fibres correlated with DE (r = 0.85) and GE (r = 0.75), both P < 0.001, n=19. r² of 0.75 is 0.56, i.e. roughly 56% of the variance in GE shared with fibre type (arithmetic on the abstract's r).
The relationship is not cycling-specific: in knee extension, % Type I also correlated with GE (r = 0.74; P < 0.001; n=13).
Authors' interpretation: Type I fibres appear substantially more efficient than Type II under these conditions, possibly through a lower rate of ATP turnover (lower VO₂ at a given power).
The classic basis for the "type I fibres → economical cyclist" narrative.

Mechanism

The abstract's proposed mechanism is a lower ATP turnover in Type I fibres at a given power. (The shortening-velocity / cadence-matching explanation often attached to this paper is not in the abstract — Inference, not tested here.)

Practical Application

Coyle et al. 1992 (n=19 highly trained cyclists): the share of Type I fibres correlated strongly with cycling efficiency (r = 0.75 for GE, 0.85 for DE); efficiency ranged 18.3-25.6% DE between riders. Baseline efficiency differs a lot between equally trained riders and much of it tracks muscle fibre type.
Not from this paper: the idea that years of endurance training shift the profile toward type I and lift efficiency is carried by REF-076 (review) and REF-077 (7-year case study).
Inference, not tested: endurance ("stayer") phenotypes with a higher type I share would carry a baseline-higher GE — a tail-wind for long-format racing.
A more recent study (Hopker 2013, REF-485) failed to reproduce the strong fibre-type ↔ GE relationship; treat REF-484 as foundational hypothesis, not closed evidence.

Limitations

n=19 — modest sample for a correlational model; men only
Single-muscle biopsy (vastus lateralis only) — does not capture whole-leg fibre profile
Cross-sectional — does not prove that training-induced shift in fibre type *causes* GE gain in the same individual
How this entry was checked and changed
Why this level

regraded 2026-09-27 from B+: cross-sectional correlational study, n=19

Content correction 2026-09-27
Level B+ -> B: cross-sectional correlational study, n=19.
"VO₂max ~71 ml/kg/min" was the abstract's body weight (71 kg); corrected to VO₂max 4.9 L/min, 71 kg.
Added the abstract's numbers (80 rpm; 52/61/71% VO₂max; Type I 32-76%; DE 18.3-25.6%; r = 0.85 DE, 0.75 GE; knee-extension r = 0.74, n=13). Removed the "60-120 rpm optimal cadence" mechanism (not in the abstract); the training-shift claim relabelled as carried by REF-076/REF-077.