Preoperative (A and C) and postoperative (B and D) photographs after endoscopic forehead-midface–lift.
In the right eye, horizontal visible iris diameter (HVID) is marked by yellow line. In the left eye, circular points mark the medial canthus (MC) and lateral canthus (LC). X represents the horizontal width between the MC and LC marked by yellow line; Y, the vertical distance between the MC and LC; and Z, the palpebral fissure width.
A minimum of 2 cm2 of periosteum in the region of the lateral canthus should remain undisturbed to prevent lateral canthal distortion. A, Undissected periosteum shaded in red. B, Corresponding soft tissue overlying preserved periosteum shaded in red.
Kolstad CK, Quatela VC. A Quantitative Analysis of Lateral Canthal Position Following Endoscopic Forehead-Midface–Lift Surgery. JAMA Facial Plast Surg. 2013;15(5):352–357. doi:10.1001/jamafacial.2013.1220
The value of this study is to evaluate outcomes of endoscopic forehead-midface–lift surgery. Many surgeons are reluctant to offer this procedure for fear of change in the shape and appearance of the eyelid.
To objectively evaluate the change in lateral canthal position following endoscopic forehead-midface–lift surgery.
A retrospective review of consecutive patients undergoing endoscopic forehead-midface–lift and lower blepharoplasty procedures for cosmetic midface rejuvenation.
A private facial plastic surgery practice.
Photometric data were obtained from before-and-after surgery images from 40 patients.
Main Outcomes and Measures
All photographs were analyzed to determine the horizontal width, vertical height, palpebral fissure width, or angle between the medial and lateral canthi. The right and left eyes were evaluated independently, with the results analyzed using a 2-tailed paired t test with a confidence interval of 0.05 or less (required for statistical significance).
The results indicated no statistically significant change in the horizontal width (right, P = .25; left P = .07), vertical height (right, P = .99; left, P = .72), palpebral fissure width (right, P = .28; left, P = .48), and angle of the lateral canthus (right, P = .99; left, P = .30) before and after surgery.
Conclusions and Relevance
The endoscopic forehead-midface–lift is a reliable method of addressing midface descent. This study objectively identified no significant differences in the position of the lateral canthus before and after surgery.
Level of Evidence
The endoscopic forehead-midface–lift is uniquely designed to address aging in the upper two-thirds of the face. This surgery combines brow-lift, midface-lift, and lower blepharoplasty procedures to resuspend ptotic tissues in a more youthful position. With advancing age, there is descent of the malar fat pad and suborbicularis oculi fat, as well as pseudoherniation of orbital fat.1,2 The combination of these events produces the characteristic double convexity and tear trough deformity of the lower eyelid and prominence of the nasolabial fold. The primary goal of a midface-lift is to reposition the suborbicularis oculi fat and malar fat pad in a superior and lateral position relative to the facial skeleton. This leads to a reduction in infraorbital hollowing, pads the infraorbital rim, and helps to correct a double-contour deformity.3,4 It also improves midface ptosis, the depth of the nasolabial fold, and lower facial jowling.5
When the midface is approached through a temporal incision, a subperiosteal dissection elevates tissue from the posterior root of the zygoma laterally to the pyriform aperture medially, with the orbital rim and lower maxilla representing the superior and inferior limits.6 After being completely released, the ptotic midface tissue is elevated to a more youthful position and secured to the deep temporal fascia. The resuspended suborbicularis oculi fat and malar fat pad lead to bunching of the skin in the lower eyelid and lateral canthal areas, requiring lower blepharoplasty and brow-lift procedures to be completed in conjunction with midface-lifts. The superolateral vector of pull created by using the deep temporal fascia as the point of fixation has led to prominent lateral canthal displacement.7- 9 Most literature citing lateral canthal distortion relies on anecdotal experience and potential risks of midface surgery. To our knowledge, only 1 published study10 has objectively evaluated the change in lateral canthal position after midface surgery.
Comparing photographs before and after surgery, Williams et al10 evaluated the change in lateral canthal position following brow-midface procedures. This well-designed study, which included 50 patients, evaluated the change in the horizontal distance and angle of inclination of the lateral canthus. The change in horizontal distance is more appropriately termed the palpebral fissure width because it comprises both horizontal and vertical components. The authors’ results indicated virtually no difference in the position of the lateral canthus following midface-lift surgery. The palpebral fissure width increased 0.09% in the left eye and 0.006% in the right eye, with neither change being statistically significant. The authors also identified a nonsignificant upward deflection of 0.77° in the left eye and a significant upward deflection of 0.86° in the right eye. They commented that no patient had objectionable canthal distortion at the end of follow-up. The relative lack of change in canthal position reported in this study would seem contrary to previously cited anecdotal evidence regarding lateral canthal distortion after midface-lifts.7- 9
A review of the statistical methods might explain this disparity. Williams et al10 analyzed the mean displacement, which may not entirely explain the true amount of lateral canthal distortion. For example, using a formula of (–25% + 25%)/2, the mean displacement of a 25% reduction and a 25% gain is 0%. A more descriptive method would be to evaluate the absolute value mean displacement, which does not average lengthenings and shortenings but rather evaluates for overall change in position. Using the previous example, the absolute value mean displacement of a 25% reduction and a 25% gain is (|–25%| + |25%|)/2 = 25% change in position.
A second limitation of the previous study10 was that the authors did not evaluate changes in the vertical axis of the lateral canthus. Instead, they used the canthal angle to describe upward and downward deflections of the lateral canthus.
The present study was designed to objectively evaluate the change in lateral canthal position following endoscopic forehead-midface–lift surgery. The position of the lateral canthus before and after surgery was evaluated in terms of change in horizontal width, vertical height, palpebral fissure width, or lateral canthal angle.
This retrospective review evaluated consecutive male and female patients undergoing endoscopic forehead-midface–lift and lower blepharoplasty procedures from 2006 through 2010 (Figure 1), all performed by the senior author (V.C.Q.). Patients were excluded if they were undergoing revision surgery, had less than 10 months of follow-up, and did not have adequate medical record documentation or photometric data.
The senior author’s endoscopic forehead-midface–lift technique, which has been described previously, will be summarized herein.3,5,11 Preoperatively, the Pitanguy line, representing the path of the frontal branch of the facial nerve, and the temporal line were marked bilaterally. The forehead and midface were injected with an equal mixture of 0.5% lidocaine with 1:200 000 epinephrine and 0.25% bupivacaine with 1:200 000 epinephrine. We made 3-cm incisions in the bilateral temples and 2 stab incisions medially. An endoscopic brow-lift periosteal elevator (Ramirez EndoForehead “T” Dissector No. 4; Snowden-Pencer; Cardinal Health) was used to elevate the superficial temporal fascia and overlying tissue off the deep temporal fascia to the temporal line. The forehead and scalp tissues were then elevated in a subperiosteal plane through the lateral incision. Under endoscopic visualization, the arcus marginalis was completely released, as was the periosteum over the glabella and nasal radix. Using the endoscope, the suprorbital and supratrochlear neurovascular bundles were dissected under visualization. If significant glabellar rhytids were present, the corrugator and procerus muscles were horizontally sectioned and cauterized.
Dissection proceeded with the endoscopic brow-lift periosteal elevator toward the zygomatic arch. The periosteum over the entire superior aspect of the arch was exposed and incised at the anterior aspect of the arch. Subperiosteal dissection then proceeded over the entire arch with the endoscopic brow-lift periosteal elevator (Ramirez EndoForehead Arcus Marginalis Dissector No. 6; Snowden-Pencer; Cardinal Health). Dissection continued on the anterior face of the maxilla toward the pyriform aperture medially and along the infraorbital rim. An important 2-cm area of periosteum around the lateral canthus was left undisturbed.
The infraorbital nerve was protected by manual pressure. Laterally, a portion of the tendinous attachments of the masseter to the zygoma was lysed. The medial and lateral subperiosteal pockets were then joined by finger dissection. The dissection was completed when there was mobility of all midface tissues, including the malar fat pad and suborbicularis oculi fat. The brow was resuspended and secured using 4-mm screws placed at the medial incisions. Before suspending the midface, a lower blepharoplasty was performed.
A total of 1 mL of 2% lidocaine with 1:100 000, mixed in equal parts with 0.25% bupivacaine hydrochloride with 1:200 000 epinephrine, was injected into the lower eyelid externally. Cotton-tipped applicators applied 4% cocaine to the conjunctiva. An additional 0.5 mL of the local anesthetic mixture was injected into the conjunctival incision and fornix.
An insulated Desmarres eyelid retractor (Bausch & Lomb) and a plastic Jaeger plate (Padgett Instruments, Integra LifeSciences) were used to protect the globe and cornea. A Colorado tip (45° angled and insulated) bovie was used to make an incision through the conjunctiva approximately 4 mm from the fornix. The incision was carried down through the lower eyelid retractors and capsulopalpebral fascia. The preseptal plane was developed by blunt dissection with cotton swabs to the infraorbital rim. Small incisions were made in the preseptal fascia with scissors and protruding fat compartments (medial, middle, and/or lateral) either excised after bipolar cautery or cauterized to ablation. Following transconjunctival removal of orbital fat, the midface was suspended.
A polyglactin 910 suture (Vicryl; Ethicon) on a UR6 needle was passed through the periosteum lateral to the zygomaticofacial foramen and tacked to the deep temporal fascia using a superolateral vector of pull. A second midfacial suspension suture was placed lateral to the Pitanguy line and secured to the deep temporal fascia. Three more suspension sutures were placed through the superficial temporal fascia just inferior to the incision and suspended higher to the deep temporal fascia to elevate the temporal tissue and prevent skin bunching from midfacial elevation.
Attention was then drawn externally, where the excess infraorbital skin was crimped using Brown-Adson forceps just below the ciliary margin, extending from 1 mm lateral to the medial canthus to beyond the lateral canthus (in a crow’s foot) by approximately 1 cm. The excess skin was pinched such that the dermatochalasis was eliminated without tension or evidence of scleral show. The pinched skin was then excised using scissors. The subcilliary incision was reapproximated using 7-0 silk suture in a running locked fashion medially and vertical mattress 6-0 nylon sutures laterally.
Photometric data were obtained from the patients’ digital images before and after surgery. Postoperative photographs required a minimum of 10 months from the date of the procedure to be evaluated. All photographs were analyzed using Adobe Photoshop’s Measure Tool (Adobe Photoshop 6.0) to determine the horizontal width, vertical height, diameter (palpebral fissure width), and angle between the medial and lateral canthi (Figure 2).
Of primary importance was eliminating any difference in camera-to-subject distance in the preoperative and postoperative photographs. Using methods devised in the Williams et al10 study, we divided the measurements between the medial and lateral canthi by a known fixed distance. The horizontal visible iris diameter (HVID) was presumed not to change between surgical procedures, using the right iris for analysis of the right eye measurements and the left iris for analysis of the left eye measurements. By converting the distance into a ratio, the change in lateral canthal position could be calculated accurately. The right and left eyes were evaluated independently, with the results analyzed using a 2-tailed paired t test with a confidence interval of 0.05 or less (required for statistical significance).
The percentage of change in the horizontal axis and palpebral fissure width was calculated by subtracting the preoperative distance (A) from the postoperative distance (B) and dividing the total by the preoperative distance. The following formula summarizes this mathematical calculation: [(B/HVIDpost) – (A/HVIDpre)]/(A/HVIDpre). This equation, however, could not be used to calculate the percentage of vertical change. When the medial and lateral canthi lay at the same vertical axis, the difference in height was 0. The change in vertical height could not be calculated with a denominator of 0. Instead, the absolute value mean change in vertical height was calculated from the absolute value mean changes in horizontal and palpebral fissure widths. Using the Pythagorean theorem, with the formula Y = √(H2 – X2), the percentage change in vertical height (Y) could be calculated by using the change in palpebral fissure distance as the hypotenuse (H) and the change in width (X).
The internal validity of the Adobe Photoshop’s Measure Tool and statistical analysis was verified by comparing the change in the left iris before and after surgery. Using the heretofore mentioned methods, we calculated a change of 1% or less (mean, 0.16%) in the diameter of the left iris before and after surgery for all eyes. The average horizontal visible iris diameter has been reported as 11.8 mm,12 with a corresponding 1% change of 0.118 mm. The medial canthus should not be affected by surgery, and all results were presumed to reflect changes in the lateral canthal position.
A total of 35 women and 5 men met inclusion and exclusion criteria for this study. Forty right eyes and 40 left eyes were evaluated independently. The mean duration between surgery and the postoperative photographs was 13 months (range, 10-19 months). The results of the statistical analysis are presented in the Table.
The mean change in the horizontal width of the right eye was –0.38%, corresponding to a shortening of the horizontal axis. The absolute value mean change was 1.83%, reflecting only the overall change without taking into account increases or decreases of distances. The mean change in horizontal distance of the left eye was 0.8%, with an absolute value mean change of 2.3%. There were no statistically significant changes in the horizontal axis of the right (P = .25) or left (P = .07) lateral canthi.
The absolute value mean change in vertical height was 0.64% in the right eye and 1.1% in the left eye. There were no statistically significant changes in the vertical axis of the right (P = .99) or left (P = .72) lateral canthi.
The mean change in the lateral canthal angle of the right eye was –0.01°, corresponding to a downward deflection of the lateral canthus. The absolute value mean change was 2.23°, which reflects only the overall change without taking into account upward or downward deflections. The mean change in the lateral canthal angle of the left eye was –0.71°, with an absolute value mean change of 2.45°. There were no statistically significant changes in the lateral canthal angles of the right (P = .99) or left (P = .30) eyes.
The mean change in the palpebral fissure width of the right and left eyes was –0.42% and 0.42%, with an absolute value mean change of 1.94% and 2.55%, respectively. There were no statistically significant changes in the palpebral fissure width of the right (P = .28) or left (P = .48) eyes.
Rejuvenation of the lower eyelid, nasojugal groove, and midface continues to inspire passionate debate. Many surgeons address prominent tear trough deformities through a lower eyelid approach with transposition of orbital and suborbicularis oculi fat, orbicularis oculi muscle plication, or other extended blepharoplasty techniques.13- 15 Proponents favoring blepharoplasty alone cite the risks of midface surgery, including lateral canthal distortion, as reasons for not pursuing a more extensive surgery.
The amount of lateral canthal displacement continues to be one of the most controversial components of midface procedures.7- 9 The present study was designed to objectively evaluate the change in lateral canthal position following endoscopic forehead-midface–lift surgery.
The results indicated no statistically significant change in the palpebral fissure width, horizontal width, vertical height, and angle of the lateral canthus following endoscopic forehead-midface–lifts. These results reflect the final position of the lateral canthus at least 10 months after surgery and do not account for transient postoperative canthal distortion. Patients are typically counseled that they can expect a temporary pull of the lateral canthus lasting 2 weeks before settling into a more natural position. Although no change was significant, a closer look at the statistical analysis reveals several important findings.
As expected, the absolute value mean displacement tended to be greater than the mean displacement. The absolute value mean displacement does not consider lengthening or shortening but reflects overall change in position. The absolute value mean change in palpebral fissure widths was 1.94% and 2.55% for the right and left eyes, respectively. A change in palpebral fissure width of 2% to 3% may be perceptible to patients. The mean displacement averages the cumulative lengthenings and shortenings of the palpebral fissure width along a single continuum. When averaging the net positive and negative changes in distance, the mean displacements were –0.42% and 0.42% for the right and left eyes, respectively. It is unlikely that changes of less than 0.5% in the palpebral fissure width would be perceptible to an observer.
The most important factor in limiting postoperative canthal distortion is conservative subperiosteal elevation over the lateral orbit and zygomaticomaxillary complex. A minimum of 2 cm2 of periosteum in the region of the lateral canthus should remain undisturbed (Figure 3). If this periosteum is completely released, the patient will be at risk for prolonged canthal tethering and superolateral displacement. By limiting the dissection around the lateral canthus, most patients can expect a temporary pull on the lateral canthus postoperatively. One week after surgery, we use massage and other conservative techniques to assist movement of the canthus to its preoperative anatomic position.
In addition to being retrospective and nonrandomized, this study had other limitations. Most notable was using ratios to evaluate the percentage of change rather than using the actual distance. A prospective study could have used a ruler or marker of known distance to be included in all before-and-after photographs. By using a known distance, the amount of canthal displacement could have been directly measured, helping to identify the amount of change in position in millimeters in addition to percentage. The preoperative and postoperative photographs ranged from 10 to 19 months after surgery. Ideally, all images would have been taken exactly at 12 months postoperatively. And finally, although the vertical position of the lateral canthus was measured directly, the absolute value mean change was not. Instead, this value was calculated using the absolute value mean changes in the horizontal axis and palpebral fissure width.
In conclusion, the endoscopic forehead-midface–lift is a reliable method of addressing midface descent. This study objectively identified no significant differences in the horizontal width, vertical height, palpebral fissure width, or angle of the lateral canthus before and after surgery.
Accepted for Publication: February 3, 2013.
Corresponding Author: Christopher K. Kolstad, MD, Department of Facial Plastic and Reconstructive Surgery, Kolstad Facial Plastic Surgery, 9850 Genesee Ave, Ste 460, La Jolla, CA 92037 (firstname.lastname@example.org).
Published Online: July 4, 2013. doi:10.1001/jamafacial.2013.1220.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: All authors.
Analysis and interpretation of data: Quatela.
Drafting of the manuscript: Kolstad.
Critical revision of the manuscript for important intellectual content: Quatela.
Statistical analysis: Kolstad.
Administrative, technical, and material support: Quatela.
Study supervision: Quatela.
Conflict of Interest Disclosures: None reported.
Correction: This article was corrected on September 19, 2013, to fix an in-text equation in the Introduction.