This document discusses techniques for using botulinum toxin type A to treat periorbital rhytids (crow's feet). It begins by providing background on the FDA approval of Botox for different conditions. It then discusses risks of off-label uses like periorbital treatment, including ptosis, ectropion, strabismus, and tearing issues. The document reviews anatomy of the orbicularis oculi muscle and its relationship to surrounding muscles. It recommends injecting the lateral aspect of the orbicularis oculi to treat crow's feet while minimizing risks.

1. SURGICAL TECHNIQUE
Treatment of Periorbital Rhytids
With Botulinum Toxin Type A
Maximizing Safety and Results
Jeffrey H. Spiegel, MD
B
otulinum toxin type A is widely used for treatment of facial rhytids, often in an “off-
label” fashion. Important mechanisms of action and safety concerns for such treat-
ments are presented along with recommendations for treatment of periorbital rhytids.
Arch Facial Plast Surg. 2005;7:198-202
Injection of botulinum toxin type A (Bo-
tox and BotoxCosmetic; Allergan Inc, Ir-
vine, Calif) for the treatment of facial
rhytids has become the most common
nonsurgical cosmetic procedure.1
The
medication has been in use for more than
30 years, and its popularity continues to
grow. Botulinum toxin type A was first ap-
proved by the Food and Drug Adminis-
tration (FDA) in 1989 to treat strabismus
and blepharospasm and was then ap-
proved in 2000 for cervical dystonia. How-
ever, for facial plastic surgeons, the most
significant milestone was probably April
15, 2002, when the FDA approved botu-
linum toxin type A for the cosmetic treat-
ment of glabellar frown lines. The FDA ac-
knowledged that clinical studies provided
to them by Allergan Inc substantiated that
the medication can safely attenuate gla-
bellar frown lines for up to 120 days.
Public safety is, of course, the primary
concern of the FDA, and they note that
medications such as botulinum toxin type
A, which are used for the treatment of con-
ditions that are not life threatening or de-
bilitating, “are subject to a greater level of
scrutiny because of the benefit-to-risk ra-
tio.”2
While the FDA has approved botu-
linum toxin type A for the treatment of
glabellar lines, and more recently for hy-
perhidrosis, the FDA remains concerned
about several ways that the drug is being
used and cautions patients that it is being
administered in health spas, gyms, hotel
rooms, at patient homes, and other
nonmedical sites and by improperly
trained practitioners. The FDA expresses
further concern about “Botox parties” be-
cause botulinum toxin type A is licensed
for marketing and distribution only in
single-use vials, and as approved should
be used within 4 hours of reconstitution.
As noted by the FDA, “Treating multiple
people with one vial violates product la-
beling, which is stated on the package in-
sert, the vial and the carton.”2
An additional area of concern for the
FDA is that physicians are using botuli-
num toxin type A for treatment of rhytids
in off-label areas such as the forehead, peri-
oral area, neck, and periorbital area. The
FDA warns,
Consumers should be aware . . . that this “off-
label” use has not been independently re-
viewed by the agency, and the safety and ef-
fectiveness of Botox injections into other
regions of the face and neck, alone or in com-
bination with the frown-lines region, have not
been clinically evaluated.2
Although not officially recognized as an
approved use by the FDA, treatment of
areas other than the glabella has become
commonplace in many practices. In fact,
detailed instructions for treatment of fore-
head rhytids, crow’s feet, nasal flare, men-
tal creasing, nasolabial folds, upper lip
rhytids, and neck rhytids are available to
anybody with access to the Internet.3
As noted by the FDA, appropriate use
of a medication such as botulinum toxin
type A is dependent on an appropriately
low risk-benefit ratio. If we assume an
overall consistent rate of systemic com-
Author Affiliations: Departments of Otolaryngology–Head and Neck Surgery and
Plastic Surgery, Boston University School of Medicine, Boston, Mass.
(REPRINTED) ARCH FACIAL PLAST SURG/VOL 7, MAY/JUNE 2005 WWW.ARCHFACIAL.COM
198
©2005 American Medical Association. All rights reserved.

2. plications (eg, headaches, flu symptoms, dysgeusia, and
dysphagia) regardless of injection area (which may not
actually be the case), then for cosmetic use, the most im-
portant consideration will be the incidence of poor aes-
thetic outcome, most commonly due to ptosis or other
undesired malposition of facial structures. In other words,
botulinum toxin type A is a reasonable treatment for
rhytids if it is effective and only rarely causes an unde-
sirable outcome.
This guideline has led many practitioners to avoid en-
tirely or to discontinue the use of botulinum toxin type
A for perioral and nasolabial fold rhytids. Of necessity,
the muscles that must be weakened to provide attenua-
tion of rhytids in these areas are those that when para-
lyzed create the undesirable appearance of facial paraly-
sis and potential functional problems with speaking or
oral sphincter competence. It is possible with finesse and
experience to improve rhytids in these areas, to narrow
an undesirably wide smile, or to reduce the alveolar show
with smiling, but the relatively high risk of a poor result
hasledmanyexperiencedaestheticsurgeonstoavoidthese
areas.
The periorbital area is different. Crow’s feet result
largely from hyperkinetic muscle function of the orbi-
cularis oculi, and experience has shown that chemode-
nervation of the lateral aspect of this muscle can greatly
attenuate the appearance of lateral orbital rhytids with-
out impairing the ability to squint and protect the eyes
from foreign bodies or bright light.
RISKS
As it is for the FDA, patient safety is the primary con-
cern of ethical physicians. Thus, for off-label applica-
tions of pharmaceuticals and other medical technolo-
gies, it is imperative that the physician be cognizant of
the potential for adverse outcomes because widely ac-
cepted treatment guidelines may not be available.
Apart from the obvious risks of bruising and bleed-
ing that result from the application of botulinum toxin
type A with a syringe and needle, the most commonly
recognized and feared complication has been upper lid
ptosis. Published reports have determined an average in-
cidence of 6.5% for this complication, although some re-
ported incidences have been much higher.4
The cause is
accepted to be contact of the levator palpebrae muscle
with the toxin, presumably through diffusion from an
appropriately distant injection site. Some researchers
believe that older patients or those with loose skin or a
weak orbital septum are particularly susceptible to this
complication, while others believe that in some cases
the problem may result from the weakening of a fronta-
lis muscle that was compensating for a preexisting de-
gree of ptosis.
In addition to upper eyelid ptosis, lower eyelid ectro-
pion can result. As with surgical treatments such as
blepharoplasty, a lower eyelid “snap-test” can provide
valuable guidance regarding the risk for development of
this complication. Similarly, in patients with a weak or-
bital septum, flaccid paralysis of the orbicularis oculi
over the lower lid can lead to pseudoherniation of infra-
orbital fat.5
Nonobstructive vision problems can also develop. The
ocular muscles insert relatively close to the anterior as-
pect of the orbit; therefore, strabismus due to lateral rec-
tus weakness can occur following treatment of crow’s feet.
Another potential complication is disruption of tear for-
mation. Whether from direct injection, diffusion into a
relevant orbital muscle, or direct involvement of the lac-
rimal gland, disruption of tear formation can occur along
with an abnormal findings on a Schirmer test.6
Cosmetic procedures are of course intended to im-
prove the patient’s appearance, and so any result that pro-
duces a less desirable appearance should rightfully be con-
sidered a significant adverse outcome. Malposition of the
eyebrow has long been recognized as a potential com-
plication of glabellar botulinum toxin type A treatment,
typically resulting in a “confused” appearance due to ex-
cessive medial elevation of the brow. Similarly, exces-
sive elevation of the lateral brow can occur following treat-
ment of lateral periorbital rhytids. This can result in an
“angry” or “sinister” appearance as the brows slope in-
feriorly toward the nose. Lateral perioral rhytids are caused
by the sphincteric action of the orbicularis oculi as it
squeezes the eyelids together. That aspect of the muscle
superior to a line through the medial and lateral canthi
logically has a downward force, which serves in part to
balance the upward pull of the frontalis thus maintain-
ing brow position. Release of this inferior force leads to
unopposed upward displacement of the lateral skin and
a superior displacement of the lateral brow. When cre-
ated as part of an intentional treatment, this upward dis-
placement can provide a nonsurgical temporal brow-lift
or what some refer to as a “chemical browlift.”7,8
Complications from treatment with botulinum toxin
type A can manifest at fairly distant sites when the drug
is used in large doses. For example, treatment of cervi-
cal dystonia has dysphagia as a recognized complica-
tion.9
However, in the appropriately low doses used for
paralysis of facial muscles, relatively distant complica-
tions can result from the close proximity of facial muscles
to one another. In the treatment of periorbital rhytids,
there is a risk for lip ptosis due to inadvertent weaken-
ing of the zygomaticus major and levator labii superi-
oris muscle groups.10
It has been suggested that this com-
plication is most likely to occur in patients who have
previously had blepharoplasty or other facial plastic sur-
gery, although these same authors would appear to re-
fute the idea of abnormal anatomy as the primary risk.
Rather, the risk is present in all patients owing to the in-
sertion of the zygomaticus major near the lateral aspect
of the orbicularis oculi.11,12
ANATOMY AND PHARMACOLOGY
Maximizing results and minimizing complications are
both achieved by a thorough understanding of how botu-
linum toxin type A works and a detailed knowledge of
the muscular anatomy in the areas being treated. Botu-
linum toxin type A is 1 of 7 known serotypes of botuli-
num toxin produced by the bacterium Clostridium botu-
linum, and 1 of 2 serotypes clinically available (A and B).
This potent neurotoxin blocks the release of acetylcho-
line at the presynaptic neuromuscular junction with re-
(REPRINTED) ARCH FACIAL PLAST SURG/VOL 7, MAY/JUNE 2005 WWW.ARCHFACIAL.COM
199
©2005 American Medical Association. All rights reserved.

3. sultant flaccid paralysis of the muscle. This stands in im-
portant contradistinction to the effect of the tetanus
neurotoxin from Clostridium tetani, which prevents the
release of neurotransmitters in the spinal cord, thus re-
sulting in spastic paralysis. Botulinum toxin (1) stays lo-
cal in the presynaptic neuromuscular junction and (2)
provides a flaccid paralysis, both of which permit its use
for the safe attenuation of rhytids.
Clearly, effective use of botulinum toxin type A re-
quires careful placement of the medication. Diffusion of
the medication can occur and depends on tissue char-
acteristics, placement, and dilution. There is little evi-
dence to suggest that the doses used for facial muscula-
ture allow for distant diffusion, although it is known that
this can occur in the higher doses used for other appli-
cations such as the treatment of muscle spasms in large
muscle groups.9,13
Exactly how far the medication in fa-
cial application diffuses is not clear: some authors sug-
gest 1 cm,14
whereas others advise that it can travel 3 cm
or more.15
These larger distances for diffusion seem un-
likely: radio-labeled botulinum toxin type A studies have
shown minimal movement. In any case, the total num-
ber of units injected in any 1 area is relatively low, and if
the medication were to diffuse over a 3-cm radius, its clini-
cal effect, if any, would probably be small.13
Additionally, authors do not agree on where to inject
the material for maximal efficacy. Some have suggested
that injection into the belly of the muscle is the best ap-
proach to minimize diffusion and maximize effect,16
whereas others propose that a subcutaneous injection is
both safe and effective.17
For treatment of periorbital rhytids, the primary muscle
of concern is the orbicularis oculi. This sphincteric muscle
sits circumferentially around the eye and has the impor-
tant function of providing deliberate closure of the eye-
lids to protect the globe. The muscle has been consid-
ered to have 3 distinct subparts: the orbital, palpebral,
and lacrimal portions. The lacrimal portion of the muscle
travels deep to the lacrimal sac and inserts on the upper
and lower eyelids at the tarsal plates. Contraction of this
part of the muscle draws the eyelids against the globe
and compresses the lacrimal sac, thus assisting tear
flow. The palpebral portion passes into the eyelid su-
perficial to the septum from the bifurcation of the me-
dial palpebral ligament to the lateral palpebral raphe.
The palpebral portion is considered to have a preseptal
and a pretarsal part, depending on which portion of the
eyelid the muscle is superficial to. Contraction of this
part of the muscle provides less forceful closure of the
eyelid, as with blinking.
The orbital portion of the muscle is that portion of the
muscle over the bony aspect of the orbit and beyond. This
muscle is near the surrounding musculature and can be
seen to blend into the frontalis, corrugator, and zygo-
maticus major muscles at its margins. In addition to pro-
viding forceful eye closure and eyebrow depression, this
aspect of the muscle creates the lateral orbital rhytids
by its pull on the overlying skin. Typically it is this area
that is treated with botulinum toxin type A, although,
the preseptal aspect of the inferior palpebral portion of
the orbicularis oculi can be treated in some situations
(Figure 1).
It is evident that the muscle groups are in continuity
with one another in these areas, and knowledge of the
specific muscle relationships may help reduce the inci-
dence of complications such as lip droop due to inad-
vertent treatment of the zygomaticus major or levator la-
bii superioris muscle groups. An analysis of the relative
muscle positions observed during cadaver dissections pro-
vides some guidance by describing specific muscle rela-
tionships in 48 facial sides.10
The relationship between the
orbicularis oculi and lip elevators was determined rela-
tive to clinically useful landmarks, including the midpu-
pillary line, the lateral canthus, and then at 1 and 2 cm
directly lateral to the lateral canthus. It was observed that
at 2 cm lateral to the lateral canthus, the orbicularis oculi
was superficial to the zygomaticus major but was only an
averageof0.5cmsuperficialtothelipelevator,withwomen
having a deeper zygomaticus major than men. At 1 cm lat-
eraltothelateralcanthus,awidelyusedlandmarkforbotu-
linum toxin type A injections in this area, the zygomati-
cus major muscle interdigitated with the orbicularis
oculi an average of 1.4 cm inferior to the Frankfort hori-
zontal line. However, in 29% of cases, the zygomaticus
was less than 1 cm from the Frankfort line in this area.
In the midpupillary dissection, the orbicularis oculi was
significantly less than 1 cm superficial to the levator la-
bii superioris.10
RECOMMENDATIONS
It would appear based on anatomic studies that many of
the conventionally touted anatomic landmarks for peri-
Figure 1. Muscular anatomy of the periorbita.
(REPRINTED) ARCH FACIAL PLAST SURG/VOL 7, MAY/JUNE 2005 WWW.ARCHFACIAL.COM
200
©2005 American Medical Association. All rights reserved.

4. orbital botulinum toxin type A administration place the
toxin in excessively close proximity to muscle groups,
which can result in undesired aesthetic outcomes. For
example, the malar eminence is frequently considered a
useful landmark for periorbital botulinum toxin type A
treatments, and the practitioner is cautioned to not go
below the malar eminence. However, in the anatomic
dissections described, the lip elevators were commonly
less than 1 cm from the orbicularis oculi at the Frank-
fort line, which is at the superior edge of the malar emi-
nence. If one accepts that botulinum toxin type A can
diffuse 1 or even 3 cm, the malar eminence is clearly too
far from the periorbital area to serve as an appropriate
guideline.
Similarly, 1 cm from the bony orbital ridge is often
considered an appropriate guideline for safety.18
This
guideline strives to avoid the complications of strabis-
mus or blepharoptosis that can occur as described by stay-
ing beyond the 1-cm accepted diffusion for botulinum
toxin. Clinical experience would suggest that this is a prac-
tical landmark, since the incidence of these complica-
tions is low. However, the zygomaticus major muscle can
be less than 1 cm from the orbicularis in areas com-
monly treated for crow’s feet, and the levator labii supe-
rioris is found less than 1 cm from the orbicularis in the
midpupillary line, as might be used to treat lower lid
rhytids. Many articles show pictures with recom-
mended injection sites too close to adjacent muscle groups
or in areas where no facial muscle is present.7,14
Careful
study of the relevant facial anatomy and relative muscle
positions will allow for a more informed choice of treat-
ment sites.
In treatment of glabellar rhytids, particularly near the
medial clubhead of the eyebrow, injection into the body
of the muscle may be the safest and most effective treat-
ment approach.19,20
However, in the treatment of peri-
orbital rhytids, a superficial injection is more appropri-
ate. Injection just deep to the dermis provides effective
treatment of those aspects of the orbital portion of the
orbicularis oculi that pull on the temporal and lateral peri-
orbital skin while maximizing distance from the sur-
rounding musculature.
Like many practitioners, I routinely dilute Botox with
4 mL of preservative-free isotonic sodium chloride (nor-
mal saline) solution into each 100-U vial for a final con-
centration of 2.5 U/0.1 mL. However, unlike many who
routinely inject 15 U or more per side, I have found that
5 to 7.5 U divided into 2 or 3 treatment sites, all more
than 1.5 cm lateral to the lateral canthus, superficial, and
superior to the Frankfort line, provides effective resolu-
tion of rhytids with no perioral or periorbital complica-
tions attributable to the toxin yet observed (Figure 2).
Some practitioners suggest reconstituting the botuli-
num toxin with either a local anesthetic with epineph-
rine, or with epinephrine added to the saline dilu-
tion.20,21
These authors propose that this further reduces
the chance of diffusion from the injection site and, if a
local anesthetic is used, may provide some immediate in-
dication of the expected results.
Other commonly touted guidelines include keeping
the patient elevated for 4 hours after treatment, recom-
mending repeated muscle contraction in the treated area,
and advising the patient not to touch the face for several
hours after treatment. While these make some intuitive
sense to minimize diffusion and target a specified muscle,
it is my belief that careful placement and avoiding over-
dilution are far more important. Thus, I do not advise
my patients of these additional restrictions and have noted
no ill effects as a result.
CONCLUSIONS
Although technically an off-label application, treatment
of lateral periorbital rhytids with botulinum toxin type
A can be safely performed if strict attention is given to
the pharmacology of reconstituted botulinum toxin and
the anatomy of periorbital muscles. Physicians who de-
sire to treat periorbital rhytids are advised to stay supe-
rior to the Frankfort line, avoid overdilution, and limit
the overall number of units of botulinum toxin injected
to this area. A superficial injection just subcutaneous pro-
vides effective treatment while minimizing exposure of
the nearby lip elevators.
Correspondence: Jeffrey H. Spiegel, MD, Departments
of Otolaryngology–Head and Neck Surgery and Plastic
Surgery, Boston University School of Medicine, 88 E New-
ton St, Suite D-616, Boston, MA 02446 (Jeffrey.Spiegel
@bmc.org).
REFERENCES
1. American Academy of Facial Plastic and Reconstructive Surgery. 2002 Mem-
bership survey: trends in facial plastic surgery [news release]. Alexandria, Va:
American Academy of Facial Plastic and Reconstructive Surgery; April 2003.
2. US Food and Drug Administration. FDA Consumer Magazine. July-August 2002..
3. Hauser RA, Wahba M. Botulinum toxin type A injections. Available at: www
.emedicine.com/plastic/topic509.htm. Accessed November 26, 2004.
4. Carruthers JA, Lowe NJ, Menter MA, et al. A multicenter, double-blinded, ran-
Figure 2. Important landmarks for treatment of lateral periorbital rhytids.
The horizontal line represents the Frankfort line; the red X is the most
prominent part of the malar eminence (note its position very inferior to the
Frankfort line). The vertical line is drawn approximately 1 cm from the lateral
canthus at the bony margin of the orbit. The green oval designates the area
for safe and effective injection of botulinum toxin type A.
(REPRINTED) ARCH FACIAL PLAST SURG/VOL 7, MAY/JUNE 2005 WWW.ARCHFACIAL.COM
201
©2005 American Medical Association. All rights reserved.

5. domized, placebo-controlled study of the efficacy and safety of botulinum toxin
type A in the treatment of glabellar lines. J Am Acad Dermatol. 2002;46:840-
849.
5. Paloma V, Samper A. A complication with aesthetic use of botulinum toxin type
A: herniation of the orbital fat. Plast Reconstr Surg. 2001;107:1315.
6. Matarasso SL. Decreased tear expression with an abnormal Schirmer’s test fol-
lowing botulinum toxin type A for the treatment of lateral canthal rhytides. Der-
matol Surg. 2002;28:149-152.
7. Ahn MS, Catten M, Maas CS. Temporal brow lift using botulinum toxin A. Plast
Reconstr Surg. 2000;105:1129-1135.
8. Frankel AS, Kamer FM. Chemical browlift. Arch Otolaryngol Head Neck Surg. 1998;
124:321-323.
9. Zesiewicz TA, Stamey W, Sullivan KL, Hauser RA. Botulinum toxin A for the treat-
ment of cervical dystonia. Expert Opin Pharmacother. 2004;5:2017-2024.
10. Spiegel JH, Derosa J. The atomic relationship between the obicularis oculi muscle
and the levator labii superioris and zygomaticus muscle complexes. Plast Re-
constr Surg. In press.
11. Matarasso SL, Matarasso A. Treatment guidelines for botulinum toxin type A for
the periocular region and a report on partial upper lip ptosis following injections
to the lateral canthal rhytids. Plast Reconstr Surg. 2001;108:208-214.
12. Fagien S. Discussion: treatment guidelines for botulinum toxin type A for the peri-
ocular region and a report on partial upper lip ptsosis following injections to the
lateral canthal rhytids by Seth L. Matarasso, M.D., and Alan Matarasso, M.D. Plast
Reconstr Surg. 2001;108:215-217.
13. Tang-Liu DD, Aoki KR, Dolly JO, et al. Intramuscular injection of 125I-
botulinumneurotoxin-complexversus125I-botulinum-freeneurotoxin:timecourse
of tissue distribution. Toxicon. 2003;42:461-469.
14. Niamtu J III. Botulinum toxin A: a review of 1,085 oral and maxillofacial patient
treatments. J Oral Maxillofac Surg. 2003;61:317-324.
15. Borodic GE, Ferrante R, Pearce LB, et al. Pharmacology and histology of the thera-
peutic application of botulinum toxin. In: Janckovic J, Hallett M, eds. Therapy
With Botulinum Toxin. New York, NY: Marcel-Dekker; 1994:119-157.
16. Carruthers J, Fagien S, Matarasso SL; Botulinum Toxin Type A Consensus Group.
Consensus recommendations on the use of botulinum toxin type A in facial
aesthetics. Plast Reconstr Surg. 2004;114(6 suppl):1S-22S.
17. Shaari CM, George E, Wu BL, Biller HF, Sanders I. Quantifying the spread of botu-
linum toxin through muscle fascia. Laryngoscope. 1991;101:960-964.
18. Carruthers JD, Carruthers A. Botulinum toxin in clinical ophthalmology. Can J
Ophthalmol. 1996;31:389-400.
19. Macdonald MR, Spiegel JH, Raven RB, Kabaker SS, Maas CS. An anatomical ap-
proach to glabellar rhytids. Arch Otolaryngol Head Neck Surg. 1998;124:1315-
1320.
20. Gassner HG, Sherris DA. Addition of an anesthetic agent to enhance the predict-
ability of the effects of botulinum toxin type A injections: a randomized con-
trolled study. Mayo Clin Proc. 2000;75:701-704.
21. Redaelli A, Forte R. Botulinum toxin dilution: our technique. J Cosmet Laser Ther.
2003;5:218-219.
(REPRINTED) ARCH FACIAL PLAST SURG/VOL 7, MAY/JUNE 2005 WWW.ARCHFACIAL.COM
202
©2005 American Medical Association. All rights reserved.