Nikos V. Margaritelis (ORCID 0000-0001-5119-427X, Aristotle University of Thessaloniki network) has 69 works, ~1,799 citations, and an h-index of 23 per OpenAlex (Semantic Scholar reports h-index 22, 1,591 citations, 63 papers) β a productive mid-career trajectory with strong output peaking 2014β2020 . His most-cited works are conceptual/methodological: Redox basis of exercise physiology (127 citations), Adaptations to endurance training depend on exercise-induced oxidative stress (117), and the NAC responder-stratification trial (112) . A distinctive theme is redox individuality β he pioneered showing that some individuals show reductive, not oxidative, stress after exercise β and falsification-oriented designs, an epistemically healthy approach rare in the field.
The reviewed work is a Fermi-style accounting model claiming ~14,570 kJ/kg (3,481 kcal/kg) central energy cost to build 1 kg of wet skeletal muscle . Strengths: transparent arithmetic, explicit falsification criteria, and honest acknowledgment that the dominant parameter (protein deposition efficiency kP = 42β52%) derives from infants and growing animals, not adult humans. Weaknesses: zero direct adult-human validation, glycogen deposition efficiency from simple accounting rather than empirical studies, unmodeled costs (satellite cells, connective tissue, inflammation, training energy), and no peer review or funding disclosure. The striking claim that common 300β500 kcal/day bulking surpluses are ~10Γ the tissue-specific requirement (~41 kcal/day) is provocative but assumes the surplus is meant purely for tissue accretion β a framing the paper itself flags as excluding training and systemic costs .
Margaritelis is a rigorous, conceptually creative researcher whose strengths are methodological self-critique (regression-to-the-mean verification, falsification designs, N-of-1 personalization) and sustained influence in redox biology. The energetics preprint exemplifies both his transparent first-principles thinking and his blind spot: elegant modeling that substitutes for the direct human measurement it acknowledges is needed. Confidence: moderate β assessment rests on citation metadata and one fully-extracted preprint.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.