Muscle soreness: what infusion experiments and biopsies actually found
Muscle soreness after unfamiliar exercise follows a different timetable from the burning or heaviness felt during a demanding effort. Experiments have approached that difference from opposite directions: putting exercise-related chemicals into resting muscle, and examining tissue after contractions that lengthen a working muscle. The first approach can generate sensations within minutes. The second follows changes over days. Together they link immediate discomfort to combinations of metabolites and delayed tenderness to the response to mechanical loading, while leaving important details of the later pain unresolved.
In a study published in Experimental Physiology in 2014, Kelly Pollak and colleagues infused combinations of lactate, ATP and protons into a small thumb muscle in ten healthy volunteers. Ultrasound guided the placement of the needles. Participants described what they felt without being supplied a menu of expected sensations. Mixtures approximating resting conditions produced almost no sensation; progressively stronger mixtures produced fatigue-related feelings and then aching or heat. The investigators also tested individual metabolites at high concentrations. Those controls did not reproduce the effects of the combination. The result supports a chemical signal involving several substances acting together, rather than lactate operating alone. It was a small infusion experiment, however, rather than a measurement of every process occurring during whole-body exercise.
The comparison with later pain comes from a different line of evidence. Jan Wilke and Michael Behringer's 2021 review describes level running that raised blood lactate without subsequent tenderness, and downhill running that caused substantial delayed muscle soreness despite low lactate. The authors also emphasize the different timing: the chemical elevation subsides soon after activity, while delayed pain emerges later. These observations place immediate metabolite measurements and later tenderness on separate physiological timetables.
The mechanical distinction behind delayed muscle soreness is active lengthening. A contracting muscle can shorten to move a load or lengthen while resisting it. In their 2001 physiological review, Uwe Proske and David Morgan described the quadriceps controlling knee bending during downhill walking as an everyday example of the latter. Their synthesis connected unfamiliar eccentric exercise with disrupted contractile structures, reduced force and altered muscle mechanics. They distinguished two proposed starting points: damage to the machinery coupling electrical excitation to contraction, and disruption of the sarcomeres that generate force. The sequence remained disputed even where structural changes were visible.
A controlled human study by Christer Malm and colleagues added a warning about interpreting biopsies. Participants performed 30 minutes of eccentric cycling, resisting a motor-driven bicycle, while a comparison group underwent repeated muscle sampling without the exercise. Blood and tissue were collected over the following week. Several markers associated with inflammatory cells increased in both groups. That comparison mattered because removing a tissue sample can itself affect the tissue subsequently examined. An immune response appearing after an exercise session was therefore not automatically a response caused by the exercise. The sampling procedure needed its own control.
The same experiment found severe muscle soreness in the exercised group, peaking at 24 or 48 hours, while the comparison group reported none. Soreness, creatine kinase activity and C-reactive protein did not track the infiltration of white blood cells into muscle. Malm's team still considered those cells important to repair, regeneration and adaptation, but its results did not support a simple chain in which the amount of inflammatory-cell invasion directly determines the amount of pain. A blood marker, a microscopic image and a person's tenderness were different measurements with different behavior.
Ji-Guo Yu and colleagues tested another part of that chain in a 2013 PLOS ONE experiment. Ten young men ran downstairs; the team examined soleus muscle samples alongside samples from six non-exercising controls. All exercised participants developed delayed tenderness, peaking between 1.5 and 2.5 days. Yet the investigators found no significant early injury to the muscle-fibre membrane or substantial inflammation using their selected stains. Fibre swelling appeared later, at 7–8 days, after the peak of discomfort. The team limited its conclusion to the muscle and exercise examined. Its findings make swelling an inadequate universal explanation for the timing of delayed muscle soreness, even though swelling can follow that exercise.
Repeated exercise supplies a further way to separate pain from other changes. In a 2019 PLOS ONE study, Zoltán Heckel and colleagues followed ten younger and ten older men through six knee-extension sessions over two weeks. They measured reported tenderness, force, electrical muscle activity and circulating muscle proteins. Soreness peaked a day after the first session in both groups, then declined toward baseline despite further sessions. Creatine kinase followed a broadly similar course. Other measurements differed: myoglobin rose selectively in the older group, and the groups showed different electrical activity during recovery. Similar soreness therefore accompanied different responses elsewhere in the measurements.
For muscle soreness, the experimental methods matter as much as the clock. Chemical infusions isolate a sensory signal; biopsy controls separate an exercise response from a sampling response; membrane stains test a proposed injury mechanism. Repeated training then tests whether tenderness follows the same course as force and blood markers. Together, these approaches require a physiological explanation to accommodate measurements that diverge. The intensity of the ache alone cannot identify the microscopic changes involved or establish that muscle function has fully recovered.
- Pollak and colleagues, Experimental Physiology (2014): Exogenously Applied Muscle Metabolites Synergistically Evoke Sensations of Muscle Fatigue and Pain in Human Subjects — pmc.ncbi.nlm.nih.gov
- Proske and Morgan, The Journal of Physiology (2001): Muscle damage from eccentric exercise — www.ncbi.nlm.nih.gov
- Malm and colleagues, The Journal of Physiology (2000): Immunological changes in human skeletal muscle and blood after eccentric exercise and multiple biopsies — pmc.ncbi.nlm.nih.gov
- Yu and colleagues, PLOS ONE (2013): Re-Evaluation of Sarcolemma Injury and Muscle Swelling in Human Skeletal Muscles after Eccentric Exercise — journals.plos.org
- Heckel and colleagues, PLOS ONE (2019): repeated knee-extension exercise in younger and older men — journals.plos.org
- Wilke and Behringer, International Journal of Molecular Sciences (2021): Is “Delayed Onset Muscle Soreness” a False Friend? The Potential Implication of the Fascial Connective Tissue in Post-Exercise Discomfort — pmc.ncbi.nlm.nih.gov