How to inject peptides — what the published protocols describe.

Search how to inject peptides and most of what comes back is somebody’s personal routine. This isn’t that. What follows is the route pharmacology recorded in published trials, plus the aseptic vial handling that keeps a reconstituted solution on specification — the two things that are actually documented.

Nearly every research peptide ships as a lyophilized powder and is reconstituted with bacteriostatic water before anything else happens. From there the literature splits three ways: into the fat layer under the skin (SubQ), into the muscle beneath it (IM), or straight into a vein (IV). Each route has its own pharmacokinetic signature, its own equipment and its own skill bar.

Two things this page is not. It is not a dosing protocol — concentrations and volumes are compound-specific, and the peptide calculator is where that arithmetic belongs. And it is not permission. Whether an unapproved compound may lawfully be given to a person is a separate question with its own answer, set out on the legal position page. Research-use-only material has no clinical pathway, which is what its label says.

SubQ: the research default

Subcutaneous injection delivers into the loose fatty layer beneath the skin. Absorption is slow, because that tissue has no dense capillary network. Trial protocols record peak plasma concentration at 1–3 hours, varying with molecule size and with any modification — acyl chains, PEGylation — engineered to slow absorption further. The depot sites those protocols specify differ from one another in absorption rate, which is why trial records name them.

SubQ is the default in published peptide research for three reasons. First, the absorption profile is forgiving: gentle peaks reduce the GI side-effect burden for compounds like the GLP-1 class, where rapid plasma spikes drive nausea. Second, the technique is teachable. It’s what insulin-dependent patients learn at home, and procedural complexity is minimal. Third, SubQ accommodates the half-life-extension chemistry that defines modern peptide therapeutics. The C20 acyl chain on tirzepatide and semaglutide is engineered specifically to bind albumin in the SubQ depot and create a slow-release reservoir.

The standard SubQ equipment is the insulin syringe. A U-100 or U-50 barrel with a fixed, short (4–8 mm), thin (29–31 gauge) needle. The short length is a safety feature: it physically can’t reach muscle through normal adult body habitus, which prevents accidental IM dosing. The thin gauge cuts injection discomfort substantially.

Site rotation is a fixture of these protocols for a pharmacokinetic reason rather than a comfort one. Repeated delivery into one subcutaneous site produces lipohypertrophy — fatty tissue thickening — and absorption from a thickened site becomes unpredictable. Decades of insulin-dependent diabetes management established the rotation convention, and the peptide trial literature inherited it.

IM: faster absorption, narrower applications

Intramuscular injection puts the dose into the muscle beneath the SubQ layer. Standard sites: deltoid (shoulder), vastus lateralis (outer thigh), or gluteal muscles. Muscle has much higher capillary density than SubQ fat, so absorption is faster — peak plasma in 30–60 minutes for most molecules.

IM shows up less often in peptide research because faster absorption isn’t usually what you want. Most peptide therapeutics are engineered for slow release, not rapid systemic exposure. IM is more common in contexts where rapid effect matters: vaccines (where adjuvant-driven local immune activation is part of the mechanism), some antibiotics, and certain hormone preparations.

Standard IM equipment is the tuberculin syringe: 1 mL barrel with a longer (16–25 mm) and wider (22–25 gauge) needle that reaches through SubQ tissue into muscle. The longer needle is what makes IM and SubQ syringes non-interchangeable.

Site selection is more constrained than for SubQ. Clinical protocols name specific anatomical landmarks — the deltoid V, the upper-outer quadrant of the gluteus medius — because the surrounding nerve and vascular structures make the margin narrow. The Z-track technique appears in the same protocols where the solution is irritating or staining: displacing tissue laterally before entry and releasing after withdrawal traps the dose in muscle. Both are clinical procedures performed by trained personnel.

IV: immediate, complete, and rarely used here

Intravenous injection delivers the full dose directly into circulation. Bioavailability: 100% instantly. Plasma concentration peaks within seconds. There’s no absorption phase — you start at distribution and elimination.

IV is the hardest of the three routes and the least common in peptide research. Most peptide therapeutics are designed for SubQ self-administration, because IV introduces real problems. It requires venous access (a skill that takes hundreds of reps to do reliably without local complications). It produces the highest possible peak concentration, which exacerbates dose-dependent side effects. The dose-to-effect relationship is harder to titrate without absorption-phase modulation.

The 2025 BPC-157 IV safety pilot is one of the rare published examples. Lee & Burgess infused two adults with 10 mg and 20 mg doses respectively, explicitly to document safety at IV doses. The authors noted no biomarker abnormalities and also that n=2 establishes nothing definitive. The choice of IV in that pilot was driven by the safety-pharmacology question; SubQ would have introduced an unwanted absorption variable.

Why SubQ dominates peptide research

The SubQ default isn’t accidental. Three design constraints push the industry toward it.

• Self-administration is the goal. Peptide therapeutics for chronic conditions (T2D, obesity, growth-hormone deficiency) require patients to self-inject for months or years. SubQ is the only route adults can reliably learn to perform.
• Slow-release chemistry needs SubQ. The C20 acyl chains on tirzepatide and semaglutide, cholesterol modifications on cagrilintide, polymer backbones on extended-release formulations — all engineered for SubQ-depot retention. IM or IV would defeat the half-life extension entirely.
• Slow absorption reduces side effects. GLP-1 GI symptoms scale with peak plasma concentration. The same dose IV would produce dramatically worse nausea than the same dose SubQ. Slow absorption is a feature, not a bug.

The 2022 SURMOUNT-1 trial of tirzepatide for obesity (Jastreboff et al., NEJM 2022) used SubQ once-weekly for all 2,539 patients. SURPASS-2 (Frías et al., NEJM 2021) used the same route. Nearly every published Phase III obesity and T2D trial in the GLP-1 / dual-agonist / triple-agonist class uses SubQ. That’s deliberate, not accidental.

Insulin syringes vs tuberculin syringes

The two syringe families aren’t interchangeable. Here’s the difference:

• Insulin syringe (U-100): 0.3 mL, 0.5 mL, or 1 mL barrel. Fixed, short (4–8 mm), thin (29–31 gauge) needle. Calibrated in “units” (1 unit = 0.01 mL). Designed for SubQ injection of small volumes. Can’t reach muscle in typical adult body habitus.
• Tuberculin syringe: 1 mL barrel with detachable needle. Length 16–25 mm, gauge 22–25. Calibrated in milliliters. Designed for IM injection or larger-volume SubQ injection.

For most lyophilized peptide doses, the insulin syringe is the right tool. Standard reconstitution concentrations produce doses in the 0.05–0.5 mL range. Insulin-syringe unit calibration gives you fine-grained measurement at that scale. Tuberculin-syringe mL calibration introduces measurement error. The exception is IM administration: the insulin syringe’s short needle can’t reach muscle, so you need a tuberculin syringe (or specialized IM syringe with detachable longer needle).

Why dose volume matters

Subcutaneous tissue absorbs only so much liquid per site before swelling and altered absorption kinetics set in. Published limits sit around 1–1.5 mL per site for adult abdomen and slightly less for thigh and upper arm, which is why protocols above that volume split the delivery.

That constraint is what drives reconstitution-concentration choices, and it runs backwards into the arithmetic. A 10 mg vial reconstituted with 2 mL of bacteriostatic water gives 5 mg/mL, so a 1 mg quantity occupies 0.2 mL — comfortably inside the limit. The same vial reconstituted with 10 mL gives 1 mg/mL, and the same 1 mg now occupies a full 1 mL, at the ceiling.

So the diluent volume chosen at reconstitution decides whether a given quantity fits a single depot at all. The peptide calculator works that relationship in both directions — vial mass and diluent volume to concentration, then concentration to draw volume in U-100 syringe units.

IM tissue can handle larger volumes (up to 2–3 mL in the deltoid, more in larger muscles), but the same site-rotation rule applies. IV is volume-unconstrained because the dose dilutes instantly into circulating blood.

Aseptic technique at the vial

This is the part of the subject that is laboratory practice rather than clinical practice, and it is the part that decides whether a reconstituted vial still meets its specification a fortnight later. Every item below is about protecting the container and the solution inside it.

• Hand hygiene before handling. Soap and water for at least 20 seconds, or a 60%+ alcohol-based hand rub. Standard bench practice, and the cheapest control available.
• Disinfect the septum every access. A single-use alcohol prep pad on the rubber stopper, for the diluent vial during reconstitution and for the reconstituted vial before every withdrawal. Let it dry: the antimicrobial action happens during evaporation, not during wet contact.
• A fresh needle for every withdrawal. A reused needle carries contamination from its last pass and has already dulled. Insulin syringes are manufactured as single-use items.
• Vary the puncture point. Repeated coring of one spot on the septum sheds elastomer into the solution and weakens the seal.
• Clear air from the barrel. Primarily a measurement question — air occupies volume that should be solution, and every concentration you derive downstream inherits the error.
• Hard-sided sharps disposal. Not household waste, and never recapped. One-handed scoop into a rigid sharps container.

The 28-day in-use window for a reconstituted multi-dose vial assumes this on every access. Septum contamination during repeated withdrawal is the dominant failure mode. The bacteriostatic agent suppresses microbial growth; it does not sterilize, and it does not recover a vial that has already been compromised.

What goes wrong at the bench

Four handling errors account for most off-specification vials, and none of them announce themselves at the time:

• Shaking instead of swirling during reconstitution. Peptides denature at the air–liquid interface, so foam is chemical damage rather than a cosmetic problem. Add the diluent down the glass wall and swirl until the cake clears.
• Reusing one syringe across withdrawals. Even inside a single session, that carries contamination back into the vial and shortens the in-use window you were relying on.
• Skipping the alcohol-dry step. Puncturing through wet alcohol discards most of the antimicrobial effect, because alcohol kills during evaporation.
• Drawing straight from the fridge. Cold solution is more viscous and shears harder through a fine needle. Protocols allow 5–10 minutes of bench warming first.

Three of the four are storage-adjacent, and the storage and stability guide covers the temperature tiers those windows come from.

Choosing the right route

Route in the published record is set by the molecule’s design and by the research question, never by preference. Tirzepatide is studied SubQ because the acylation chemistry was engineered for a subcutaneous depot. BPC-157 appears across intraperitoneal, intragastric, intra-articular, intravesical and IV work, because the injury model dictated the route each time.

The practical consequence is a limit on what any of this evidence covers. A safety and efficacy profile attaches to the route it was generated in: SURMOUNT is a subcutaneous readout, so it is evidence about subcutaneous administration and nothing else. Read across to a different route and there is no human study behind the number any more — the pharmacokinetics diverge and the characterization simply does not exist.