Introduction
The glabellar complex remains the most frequently treated region in aesthetic practice and the region about which practitioners most often ask the wrong question. That question is almost always about units. How many units for a strong frown, how many for a man, how many to make the result last past Christmas. The dose is the easiest variable to change and therefore the one clinicians reach for first, which is precisely why the glabella continues to generate a disproportionate share of both dissatisfied patients and reported complications.
The evidence base does not support dose as the principal determinant of outcome. It supports anatomy, pattern recognition, injection plane and entry angle as the principal determinants, with dose functioning as a coarse adjustment applied after those decisions have been made correctly. A twenty-unit treatment placed accurately outperforms a forty-unit treatment placed carelessly, and it does so at a fraction of the risk. This review sets out the anatomical basis for that claim, examines what the trial data actually show about the relationship between dose, duration and adverse events, and translates both into a defensible dosing framework.
Throughout, doses are expressed in onabotulinumtoxinA-equivalent units unless otherwise stated. Readers should note that a substantial proportion of what follows describes off-label practice. The licensed indication is narrow; competent practice around it is not, but it carries a correspondingly higher documentation and consent burden.
The Glabellar Complex Is a Five-Muscle System
Most training describes the glabella as procerus plus two corrugators. That description is adequate for the licensed pattern and inadequate for anything else. The functional unit comprises five muscles: procerus in the midline, the paired corrugator supercilii, the paired depressor supercilii, and the medial fibres of orbicularis oculi. All five share a depressing vector on the medial brow, and all five contribute to the visible frown in proportions that vary considerably between individuals.
Procerus arises from the fascia over the lower nasal bone and inserts into the skin of the lower forehead between the brows. Its contraction produces the horizontal root-of-nose line and pulls the medial brow inferiorly. Corrugator supercilii arises from the superomedial orbital rim and runs obliquely upwards and laterally, interdigitating with frontalis and orbicularis before inserting into the dermis of the mid-brow. Its contraction draws the brow medially and downward, producing the vertical lines. Depressor supercilii, frequently treated as a corrugator variant, arises from the frontal process of the maxilla near the medial canthal tendon and depresses the medial brow head directly.
The clinical consequence of this arrangement is that residual medial brow depression after an apparently adequate treatment is usually depressor supercilii or medial orbicularis activity rather than an under-dosed corrugator. Adding units to the corrugator belly in that situation does not resolve the problem and moves the treatment closer to the risk threshold for no therapeutic gain.
When the vertical lines soften but the medial brow still sits low and heavy on animation, the unaddressed muscle is almost always depressor supercilii. The correction is one to two units placed medially at the brow head, not a larger corrugator dose.
Standard Labelled Doses
Licensed glabellar doses were established through pivotal trials designed to demonstrate efficacy against placebo at a fixed dose, not to identify an optimal dose for an individual. They should be read as a validated starting point rather than a ceiling or a target.
| Product | Dose (Units) | Typical Duration |
|---|---|---|
| Botox (onabotulinumtoxinA) | 20 U | 3–4 months |
| Dysport (abobotulinumtoxinA) | 50 U | 3–4 months |
| Xeomin (incobotulinumtoxinA) | 20 U | 3–4 months |
| Jeuveau (prabotulinumtoxinA) | 20 U | 3–4 months |
| Daxxify (daxibotulinumtoxinA) | 40 U | 5–6 months |
Product units are not interchangeable. Dysport requires a conversion factor of approximately 2.5:1 to 3:1 against onabotulinumtoxinA; onabotulinumtoxinA, incobotulinumtoxinA and prabotulinumtoxinA can be treated as broadly unit-equivalent for glabellar purposes. DaxibotulinumtoxinA achieves its extended duration through a peptide excipient that increases residence at the neuromuscular junction rather than through a larger effective dose, and its units should not be converted using a simple ratio.
Off-Label Dose Ranges in Practice
Light dosing
Partial softening with retained expression. Suited to first treatments, younger patients with dynamic-only lines, and those whose primary anxiety is looking treated.
- • Duration typically 2–3 months
- • Very low adverse-event rate
- • Higher rate of two-week top-up
Standard dosing
The dose validated in pivotal trials and the defensible default for a new patient of average muscle bulk. Most patients reach none-or-mild severity at week four.
- • Duration typically 3–4 months
- • Adverse events below 5 per cent
- • Widest supporting evidence
High dosing
Reserved for documented inadequate response at standard dose with correct technique, or for markedly hypertrophic corrugators, most often in men.
- • Modest duration gain
- • Brow heaviness becomes common
- • Greater cumulative antigenic load
Contraction Pattern Determines Distribution
Before any dose is chosen, the patient should be asked to frown maximally and the pattern of contraction observed and recorded. Five reproducible patterns are described, and each redistributes the same total dose differently. Pattern recognition is the single most useful refinement available to a practitioner who is already dosing correctly, because it improves the result without increasing the quantity of toxin injected.
U pattern
Corrugator-dominant, procerus quiet
V pattern
Balanced corrugator and procerus
Omega
Strong procerus with medial orbicularis
Converging arrows
Deep medial corrugator vector
Inverted omega
Depressor supercilii dominant
A U pattern indicates corrugator dominance with a relatively quiet procerus; weighting should shift laterally along the corrugator with a reduced midline dose. A V pattern reflects balanced activity and is well served by the classic distribution. An omega pattern, with strong procerus recruitment and visible medial orbicularis involvement, requires greater midline weighting and often a small medial brow-head point. Converging arrows indicate a deep medial corrugator vector and reward increased depth at the medial points rather than increased volume. The inverted omega reflects depressor supercilii dominance and is the pattern most often mistreated by escalating corrugator dose.
Photograph the maximal frown at every first treatment. Pattern is stable over time in most patients, which makes the baseline image the most useful single item in the record when a later treatment produces an unexpected result.
Injection Pattern, Depth and Entry Angle
The licensed five-point pattern delivers twenty units across one midline and four paired points. Depth alone is an incomplete instruction: every stated depth must be paired with an entry angle and a vector, because the same depth reached from a different angle places toxin in a different plane and carries a different diffusion risk.
The procerus point is entered perpendicular to the skin, in the midline just above the nasofrontal angle, and advanced to periosteum before withdrawing marginally. Medial corrugator points are also entered perpendicular, deep into the muscle belly where it lies against bone. Lateral corrugator points are the exception: they should be entered at approximately forty-five degrees, bevel up, with the needle directed medially and superiorly so that the depot is placed away from the orbital rim rather than towards it.
Two geometric rules govern safety in this region. All injections should sit at least one centimetre above the bony orbital rim, and lateral points should remain medial to the mid-pupillary line. Injections that violate either rule increase the probability that toxin will track through the orbital septum, irrespective of how conservative the total dose was.
Dilution and Volume
Volume is an independent variable that is frequently ignored. A 100-unit vial reconstituted in 2.5 ml delivers four units per 0.1 ml; the same vial in 4 ml delivers 2.5 units per 0.1 ml. The second preparation places sixty per cent more fluid at each point for the same unit dose, and that additional volume increases the field of diffusion. In the glabella, where the margin between the corrugator and the levator is measured in millimetres, higher dilutions should be treated as a deliberate choice made for a reason and not as a default. Concentrated preparations, typically 2 to 2.5 ml per 100-unit vial, are preferable where precision matters more than spread.
Duration of Effect: What the Evidence Shows
| Product/Dose | Median Duration | Reported Range |
|---|---|---|
| Onabotulinum, inco-, prabotulinum (20 U) | 3–4 months | 2.5–5 months |
| Abobotulinum (50 U) | 3–4 months | 2.5–5 months |
| Daxibotulinum (40 U) | 5–6 months | 4–8 months |
Duration is not a linear function of dose. Increasing from ten to twenty units produces a clinically meaningful extension. Increasing from twenty-five to fifty units produces perhaps two to four additional weeks in exchange for a threefold or greater increase in adverse-event probability. The curves diverge, and the point at which they diverge defines the clinically efficient dose region.
If a patient reports that treatment "wears off in eight weeks", interrogate technique and pattern before increasing the dose. True early loss of effect at an accurate twenty-unit treatment is uncommon; misplaced depots, over-dilution and unrecognised depressor supercilii activity are all far more common explanations.
Dose vs Duration vs Side Effects
| Dose Strategy | Expected Duration | Side-Effect Risk | Appropriate Candidates |
|---|---|---|---|
| Light (10–15 U) | 2–3 months | Very low (<2%) | First treatment, dynamic-only lines, expression-preserving request |
| Standard (20 U) | 3–4 months | Low (<5%) | Default for most adults of average muscle bulk |
| High (30–40 U) | 4–5 months | Moderate (5–10%) | Documented inadequate response, hypertrophic corrugators |
| Very high (>40 U) | 4–6 months | Higher (10–15%) | Exceptional cases with explicit documented rationale |
Practitioner Survey Data
Survey responses from 243 practising aesthetic physicians describe real-world dosing preferences. Survey data reflect practice rather than evidence, but they are useful for showing where consensus sits and where it diverges from the trial literature.
Preferred starting doses
Adjustment by sex
The majority position — a fifteen to twenty unit start with upward adjustment for greater muscle bulk — aligns closely with the trial evidence. The minority who begin routinely at twenty-five to thirty units are operating at the shoulder of the risk curve for a duration gain that the data do not reliably support.
Safety Profile and Adverse Events
The safety profile of glabellar botulinum toxin is excellent when technique is sound. Almost all reported complications are attributable to plane, position or volume rather than to total units.
<5% incidenceCommon
- ○Injection-site bruising or swelling
- ○Headache, usually mild and self-limiting within 24–48 hours
- ○Transient sensation of brow heaviness
<1% incidenceSerious
- ●Eyelid ptosis (blepharoptosis)
- ●Brow ptosis
- ●Diplopia
- ●Asymmetry requiring correction
The Mechanism of Eyelid Ptosis
Blepharoptosis follows diffusion of toxin through the orbital septum to levator palpebrae superioris. Three technical factors make it more likely: injection placed too close to the orbital rim, injection placed too superficially so that toxin sits in a plane continuous with the septum, and excessive volume at a single point. None of these is a function of total dose, which is why ptosis is reported after conservative treatments delivered badly and avoided after larger treatments delivered accurately.
Established ptosis is managed expectantly with reassurance and a clear timeline, since resolution follows the pharmacological course of the toxin over four to twelve weeks. Apraclonidine 0.5 per cent drops stimulate Müller's muscle and typically raise the lid by one to two millimetres, providing cosmetic mitigation while recovery occurs. Brow ptosis, by contrast, usually reflects over-treatment of frontalis rather than the glabella and is prevented by respecting the lower limit of frontalis dosing in patients who recruit it compensatorily.
Brow shape change deserves separate mention. Excessive medial corrugator treatment without corresponding attention to lateral frontalis can produce an over-arched, quizzical brow. This is not a complication in the pharmacological sense; it is a distribution error, corrected by one to two units placed laterally in frontalis at the two-week review.
Special Considerations
Male patients
Greater corrugator mass and a lower, flatter brow position mean men commonly require 25–30 per cent higher doses. The brow must remain flat: over-arching is more conspicuous and less acceptable in male patients.
Treatment history
Long-term regular treatment produces genuine muscle atrophy, and dose requirements often fall over successive years. Reassessing rather than repeating the previous prescription is appropriate at approximately the fourth treatment.
Static lines
Deeply etched static lines will not resolve with toxin alone at any dose. They require dermal remodelling alongside relaxation, and this must be stated before treatment rather than explained afterwards.
The Apparent Non-Responder
True immunological non-response, mediated by neutralising antibodies to the neurotoxin complex, is rare in aesthetic dosing. Before it is invoked, four commoner explanations should be excluded: injection into the wrong plane, an unrecognised contraction pattern with untreated depressor supercilii activity, over-dilution reducing the concentration delivered to the target, and product handling or storage error. Where genuine secondary non-response is suspected, minimising cumulative dose, extending inter-treatment intervals beyond twelve weeks, avoiding top-ups within the interval, and considering a formulation without complexing proteins are the recognised mitigations. Routine dose escalation is the practice most likely to create the problem it is intended to solve.
Risk Mitigation and Aftercare
Record the maximal-frown pattern, brow position at rest, and any pre-existing lid or brow asymmetry with photography.
One centimetre above the orbital rim; lateral points medial to the mid-pupillary line; no exceptions for a patient in a hurry.
Concentrated reconstitution, small depots, and no more than 0.1 ml at any single glabellar point.
Twenty units for a new adult patient, adjusted at review on the basis of observed response rather than anticipation.
The point at which effect is fully established, asymmetry can be corrected with one or two units, and the record for the next cycle is made.
Post-treatment advice should be proportionate to the evidence supporting it. Avoiding rubbing or massaging the treated area for the remainder of the day is sensible on diffusion grounds. Remaining upright for four hours and avoiding strenuous exercise on the day of treatment are conventional and low-cost, though the supporting evidence is weak. Elaborate multi-day restriction lists are not evidence-based and tend to reduce rather than increase patient confidence.
Clinical Recommendations
- Assess first. Contraction pattern, brow position and muscle bulk determine distribution; dose is chosen afterwards.
- Default to 20 units. Onabotulinumtoxin-equivalent, five-point pattern, for a new adult patient of average bulk.
- Reduce to 15 units for a first treatment where the patient's stated priority is retained expression.
- Escalate only on evidence of inadequate response at an accurately delivered standard dose, and document the rationale.
- Treat depth and angle as prescribed variables recorded in the notes alongside units.
- Review at two weeks and correct with one to two units rather than re-treating the whole complex.
Conclusion
The evidence supports the licensed twenty-unit onabotulinumtoxinA-equivalent dose as the appropriate default for most patients seeking glabellar treatment. It produces reliable efficacy for three to four months with an adverse-event rate below five per cent, and it does so across a wide range of facial morphologies.
Higher doses extend duration modestly and increase risk substantially. The clinically efficient region lies between fifteen and twenty-five units, and practitioners who work outside it should be able to state why in the record. Where results disappoint, the productive response is to re-examine the contraction pattern, the injection plane and the entry angle before reaching for more units, because those three variables account for most of the difference between an adequate glabellar treatment and an excellent one.
Glabellar treatment is a small technical procedure performed close to structures that do not forgive imprecision. Treating it as a dosing exercise rather than an anatomical one is the most common error in aesthetic practice, and it is entirely avoidable.
References
- Carruthers JA, Lowe NJ, Menter MA, et al. A multicenter, double-blind, randomized, placebo-controlled study of the efficacy and safety of botulinum toxin type A in the treatment of glabellar lines. J Am Acad Dermatol. 2002;46(6):840–849. PubMed
- Carruthers A, Carruthers J, Said S. Dose-ranging study of botulinum toxin type A in the treatment of glabellar rhytids in females. Dermatol Surg. 2005;31(4):414–422. PubMed
- Carruthers A, Carruthers J. Prospective, double-blind, randomized, parallel-group, dose-ranging study of botulinum toxin type A in men with glabellar rhytids. Dermatol Surg. 2005;31(10):1297–1303. PubMed
- Cook BE, Lucarelli MJ, Lemke BN. Depressor supercilii muscle: anatomy, histology and cosmetic implications. Ophthalmic Plast Reconstr Surg. 2001;17(6):404–411. PubMed
- de Almeida ART, da Costa Marques ER, Banegas R, Kadunc BV. Glabellar contraction patterns: a tool to optimize botulinum toxin treatment. Dermatol Surg. 2012;38(9):1506–1515. PubMed
- Hexsel D, Brum C, Siega C, et al. Evaluation of the spread of botulinum toxin type A according to reconstitution volume. J Drugs Dermatol. 2012;11(6):e33–e37. PubMed
- Bertucci V, Solish N, Kaufman-Janette J, et al. DaxibotulinumtoxinA for injection has a prolonged duration of response in the treatment of glabellar lines (SAKURA 1 and 2). J Am Acad Dermatol. 2020;82(4):838–845. PubMed
- Naumann M, Jankovic J. Safety of botulinum toxin type A: a systematic review and meta-analysis. Curr Med Res Opin. 2004;20(7):981–990. PubMed
- Bellows S, Jankovic J. Immunogenicity associated with botulinum toxin treatment. Toxins. 2019;11(9):491. PMC
- Cochrane Library. Botulinum toxin for the treatment of facial lines and wrinkles — systematic reviews. Cochrane Library