BPC-157 for Muscle Healing — Top 10 Animal Studies Reviewed (2026 Evidence-First Breakdown)

Informational only. Not medical advice. BPC-157 is not FDA-approved for any clinical use. Most published evidence is animal or preclinical, and there are no human muscle-healing trials. Do not start, stop, or change any medical treatment based on what you read here. Speak with a licensed clinician before considering any peptide.

People search for "BPC-157 for muscle healing" and find dosing protocols. The real data is narrower. Almost every study below was done in rats by Predrag Sikiric's lab in Croatia.

This article reviews 10 of the most-cited peer-reviewed papers on BPC-157 in muscle injury models. Each entry names the model, the sample, the outcome, and the year. Read it as a map of what was tested, not a guide to dosing.

For the broader evidence base across tendon, gut, brain, and human safety, see our top 10 BPC-157 research studies review.

Does BPC-157 actually help muscle injuries heal faster?

In rats, yes — across every model tested. In humans, no published evidence exists. Across crush, transection, junction cut, detachment, denervation, and toxin models, treated rats heal faster than untreated controls. Effect sizes are large enough that walking gait, biomechanical strength, and histology often return to near-baseline by day 14-42.

The catch: 35 of the 36 studies in the 2025 systematic review were preclinical. The only clinical paper was a knee-pain case series in 12 patients — not a muscle injury trial (Vasireddi et al., 2025).

Translation from rat to human muscle physiology is not guaranteed. The rat data is consistent in direction; the human data does not exist.

What's the strongest animal evidence for BPC-157 muscle healing?

The 2008 gastrocnemius crush paper by Mihovil and Staresinic is the most-cited acute trauma model. It crushed rat calf muscle at 0.727 Ns/cm² force, dosed BPC-157 for 14 days, and measured creatine kinase, lactate dehydrogenase, contracture, edema, and walking gait. Treated rats had biochemical and functional recovery by day 14 (Mihovil/Staresinic et al., 2008).

The 2021 Krezic/Japjec myotendinous junction paper is the strongest model on selection grounds: untreated rats with a cut junction do not heal spontaneously, so the BPC-157 treatment effect is not confused with natural healing. Treated junctions reformed by day 28-42 (Krezic et al., 2021).

The 2025 Strbe/Drinkovic detachment paper added ultrasound and MRI endpoints on a 90-day follow-up — the longest imaging arc in the muscle literature to date (Strbe et al., 2025).

Has BPC-157 been tested in humans for muscle injury?

No. As of May 2026, zero published randomized controlled trials of BPC-157 in human muscle injury exist. The 2025 systematic review (Vasireddi et al.) screened 544 papers and found exactly one clinical study — a 12-patient knee-pain case series, not a muscle injury trial.

A separate 2025 IV safety pilot in 2 healthy adults found no acute adverse events but did not measure efficacy (Lee et al., 2025).

Athletes use BPC-157 widely off-label. Published case reports tied to muscle healing are mostly absent in 2026. The World Anti-Doping Agency lists BPC-157 as prohibited in sport at all times.

The 10-study summary table

<a id="study-table"></a>

#	Study (lead author, year)	Model	Primary outcome
1	Mihovil et al., 2008 — gastrocnemius crush	Rat	Less hematoma, restored walking by day 14
2	Pevec et al., 2010 — muscle + corticosteroid	Rat	Healing restored despite methylprednisolone
3	Staresinic et al., 2006 — quadriceps tendon	Rat	Biomechanical strength restored
4	Krezic / Japjec et al., 2021 — myotendinous junction	Rat	Junction reformed by day 28-42
5	Sikiric et al., 2014 — rotator cuff tear	Rat	Mobility and muscle strength matched healthy
6	Strbe / Drinkovic et al., 2025 — quadriceps detachment	Rat	Muscle-to-bone reattachment, walking restored
7	Staresinic et al., 2022 — denervation atrophy	Rat	Atrophy prevented over 1-year follow-up
8	Brcic et al., 2009 — angiogenesis in muscle	Rat	VEGFR2 upregulated at injury site
9	Sikiric et al., 2016 — succinylcholine paralysis	Rat	Leg contracture and twitches eliminated
10	Vasireddi et al., 2025 — systematic review	Meta	35 preclinical, 1 small human, no muscle RCTs

The 10 most-cited muscle studies

1. Mihovil / Staresinic et al. (2008) — gastrocnemius crush in rats

Best for: the most-cited acute soft-tissue trauma paper Model: rat gastrocnemius crush, force 0.727 Ns/cm², dosed 14 days Standout finding: less hematoma, less edema, restored walking by day 14

This paper crushed rat calf muscle. BPC-157 was given as a shot or a topical cream. Treated rats had lower creatine kinase, less leg contracture, and "almost no scar tissue" (Staresinic et al., 2008).

It is the most-cited paper clinicians point to when they suggest BPC-157 off-label for sports injuries. The data is rat-only.

Strengths

• Both biochemical and functional endpoints
• Two delivery routes (intraperitoneal and topical) compared

Limitations

• Rat crush model does not equal a human contact sport injury
• No human replication exists in 2026

2. Pevec et al. (2010) — muscle healing despite corticosteroid

Best for: evidence in a drug-impaired model Model: rat gastrocnemius crush plus methylprednisolone Standout finding: BPC-157 restored muscle healing despite systemic steroid

Steroids slow muscle repair. This paper injured the gastrocnemius, gave 6α-methylprednisolone, then dosed BPC-157. The steroid-only group healed poorly, while the BPC-157 plus steroid group regained function (Pevec et al., 2010).

It is one of the few BPC-157 papers that uses an active blocking agent, not just vehicle.

Strengths

• Active comparator arm (steroid-impaired healing)
• Endpoints across days 1, 2, 4, 7, and 14

Limitations

• One lab, one rat strain
• No human steroid-induced myopathy data exists

3. Staresinic et al. (2006) — quadriceps tendon transection

Best for: the foundational tendon-muscle paper Model: rat quadriceps tendon transection, n=120 Standout finding: biomechanical strength and gait restored vs untreated

The quadriceps tendon was cut away from the muscle. BPC-157 was given as a shot or as a cream. By day 14, treated rats walked with less limp and their tendons held more pull force (Staresinic et al., 2006).

The paper seeded the wider claim that BPC-157 helps myotendinous junction repair. It is rat-only.

Strengths

• Large sample size for an animal study
• Function, biomechanics, and histology all measured

Limitations

• No follow-up beyond 14 days
• Rat tendon biology does not translate cleanly to human

4. Krezic / Japjec et al. (2021) — disabled myotendinous junction

Best for: the only published rat myotendinous junction model Model: quadriceps tendon dissected from muscle, n=multi-arm Standout finding: junction reformed by day 28 and 42 vs no spontaneous healing in controls

Untreated rats with a cut myotendinous junction did not heal. Controls showed atrophy and a lasting defect. BPC-157 given as a shot or in drinking water produced near-normal junction structure on histology (Krezic et al., 2021).

A 2018 FASEB Journal abstract reported the early version of this result (Japjec et al., 2018).

Strengths

• A model that does not heal on its own — clear effect size
• Multiple dose levels and routes tested

Limitations

• One Zagreb lab, no independent replication
• Rat anatomy at the junction differs from human

5. Sikiric et al. (2014) — rotator cuff tear in rats

Best for: the only published BPC-157 rotator cuff paper Model: rat supraspinatus + infraspinatus detachment, n=48 Standout finding: mobility range and muscle strength approached healthy controls

Researchers detached the supraspinatus and infraspinatus, then dosed half the rats with BPC-157 and half with saline. Treated rats recovered shoulder mobility close to baseline, while controls had less range and weaker muscle (Sikiric et al., 2014).

This is a FASEB Journal meeting abstract. It is not a full paper, and the full data was never published.

Strengths

• Clinically relevant injury model
• Functional recovery measured, not only histology

Limitations

• Abstract only, no peer-reviewed full text
• n=48 in a single lab

6. Strbe / Drinkovic et al. (2025) — quadriceps muscle-to-bone reattachment

Best for: the most recent rat muscle paper Model: rat complete and partial quadriceps detachment from bone Standout finding: walking pattern fully restored, muscle approached bone on ultrasound

This 2025 paper detached the rectus and vastus muscles from the bone. Untreated rats had locked knees and a bad gait out to day 90. Oral BPC-157 produced gross, ultrasound, MRI, and biomechanical recovery at every timepoint (Strbe et al., 2025).

It is one of the few muscle papers that uses imaging endpoints, not just histology.

Strengths

• Ultrasound and MRI endpoints
• 90-day follow-up window

Limitations

• Still a rat model, still one lab
• Oral dosing pharmacokinetics in humans are unstudied

7. Staresinic / Sikiric et al. (2022) — denervation atrophy in rats

Best for: chronic-denervation data Model: rat gracilis muscle denervation, 1-year follow-up Standout finding: atrophy prevented and function preserved at one year

Denervated rat gracilis muscle normally shrinks. Fiber size drops, weight drops, odd nuclei appear. BPC-157 across many dose levels reversed all of those changes, and treated muscle looked like the healthy leg on microscopy (Staresinic et al., 2022).

This is a narrative-plus-data review on striated, smooth, and heart muscle. It pulls many Zagreb experiments under one roof.

Strengths

• One-year follow-up is unusually long for the BPC-157 literature
• Consistent finding across multiple dosing regimens

Limitations

• Review structure makes it hard to isolate one experiment
• No human nerve injury or atrophy data exists

8. Brcic et al. (2009) — angiogenesis in muscle and tendon

Best for: mechanism evidence in muscle tissue Model: rat muscle and tendon injury, histology + VEGFR2 staining Standout finding: VEGFR2 expression upregulated at the injury site

Brcic and team looked at how BPC-157 changes blood vessel growth during muscle healing. Treated tissue showed more microvessels and higher VEGFR2 staining than controls (Brcic et al., 2009).

A 2017 mechanism paper extended this to the VEGFR2-Akt-eNOS pathway in cell culture (Hsieh et al., 2017).

Strengths

• Maps a plausible molecular pathway for muscle effects
• Replicates angiogenesis findings from skin-wound studies

Limitations

• Pathway activation does not equal clinical benefit
• No human pharmacokinetic data confirms exposure

9. Sikiric et al. (2016) — succinylcholine paralysis and muscle damage

Best for: acute muscle-toxin reversal data Model: rat anterior tibial muscle, succinylcholine 1.0 mg/kg Standout finding: BPC-157 eliminated leg contracture and post-paralytic hyperalgesia

Succinylcholine causes muscle twitching and paralysis. This paper gave it to rats, then dosed BPC-157. Treated rats had no leg lock-up, less swelling, and kept their diaphragm fiber count (Sikiric et al., 2016).

The paper sits next to a companion bupivacaine study from the same year (Boban Blagaic et al., 2016).

Strengths

• Acute pharmacologic injury, well-defined endpoints
• Endpoints at 3, 30 min, and 1, 3, 5, 7 days

Limitations

• Toxin reversal does not equal sports injury healing
• No human anesthetic complication trial exists

10. Vasireddi et al. (2025) — systematic review of BPC-157 in sports medicine

Best for: the only systematic review covering muscle endpoints Model: meta-review of 36 studies (35 preclinical, 1 clinical) Standout finding: zero published randomized controlled trials in human muscle injury

This 2025 review searched PubMed, Cochrane, and Embase up to June 2024. Of 544 papers screened, 36 made the cut. 35 were preclinical, and one was a small knee-pain case series in 12 patients (Vasireddi et al., 2025).

The authors note BPC-157 lacks FDA approval and is banned in pro sport, despite rising use by athletes and clinics.

Strengths

• Pre-registered systematic methodology
• Captures the full clinical evidence gap in one paper

Limitations

• Confirms what the rest of this list already shows — no human muscle RCTs
• Cannot fix the underlying single-lab dominance in the preclinical record

Bottom line

The BPC-157 muscle-healing record is steady in direction and narrow in source. Across crush, transection, junction cut, detachment, denervation, and toxin models, treated rats heal faster than controls. Almost every paper traces back to the Sikiric lab in Zagreb.

This is not informed consent for self-injection. As of May 2026, zero human trials of BPC-157 for muscle injury have been published. The FDA has not approved BPC-157 for any use, and WADA bans it in sport at all times (Vasireddi et al., 2025).

Related Reading

• Top 10 BPC-157 research studies (2026) — broader evidence across tendon, gut, brain, and human safety
• BPC-157 for tendon healing: animal study review
• Compounded peptides vs research chemicals: 10 key differences (2026)

Frequently asked questions

Why are there no human muscle trials of BPC-157? No sponsor has funded a randomized trial in human muscle injury. The 2025 systematic review found 35 preclinical studies and one small knee-pain series, with no muscle RCT (Vasireddi et al., 2025).

Are there human cases of BPC-157 used for muscle injury? Athletes use it widely off-label. But published case reports tied to muscle healing are mostly absent in 2026. The one clinical paper covered knee pain, not muscle injury.

Why is most BPC-157 muscle research from one lab? The Zagreb group led by Predrag Sikiric has run most BPC-157 work since the 1990s. Outside replication is thin, which is a known concern in the evidence base.

Does BPC-157 work in humans the way it works in rats? Unknown. Animal data does not always translate. A 2025 IV safety pilot in two adults showed no acute events but did not measure efficacy.

Is BPC-157 legal to use for muscle recovery? It is sold online as "research use only." It is not legal to compound for human use in the US. WADA lists it as banned in sport at all times.

---

Researched and drafted by Theo Park, an AI editorial persona at Peptide Front, against published sources. Reviewed by our editorial team.