At time 0, the suction pressure is applied to the skin. At this point, the device begins to measure the distance the skin deforms over time. The Cutometer provides the elastic resistance (Ue) and viscoelastic resistance (Uv). At 3 seconds, the Cutometer suction pressure is removed. At this point, the maximum deformation of the skin is measured (overall pliability [Uf]). The skin then recoils, and the Cutometer measures this change. At the end of the 3-second period of no suction, the initial recoil elastic (Ur) and the total elastic recoil (Ua) are measured.
Error bars indicate 95% CI.
Bonaparte JP, Ellis D. Alterations in the Elasticity, Pliability, and Viscoelastic Properties of Facial Skin After Injection of Onabotulinum Toxin A. JAMA Facial Plast Surg. 2015;17(4):256–263. doi:10.1001/jamafacial.2015.0376
This prospective cohort study provides evidence and information on the mechanism of action of onabotulinum toxin A on the reduction of skin elasticity and pliability. Understanding the natural course that onabotulinum toxin A has on the elasticity of skin may help physicians understand why there appears to be a progressive reduction in wrinkle levels with repeated treatments
To determine whether onabotulinum toxin A increases skin pliability and elasticity with a corresponding decrease in the contribution of the viscoelastic component of skin resistance.
Design, Setting, and Participants
From October 1, 2012, through June 31, 2013, this prospective cohort study enrolled 48 onabotulinum toxin A–naive women (mean [SD] age, 55.2 [11.3] years) with a minimum of mild wrinkle levels at the glabella and lateral orbit (43 completed the study). Patients were treated at a private cosmetic surgery clinic with onabotulinum toxin A and assessed at baseline and 2 weeks, 2 months, 3 months, and 4 months after injection.
Standardized onabotulinum toxin A was administered to patients’ glabella, supraorbit, and lateral orbit.
Main Outcomes and Measures
Skin pliability, elastic recoil, and the ratio of viscoelastic resistance (Uv) to elastic resistance (Ue).
For the supraorbit, there was a significant effect of time on pliability (F = 20.5), elastic recoil (F = 6.92), and Uv/Ue ratio (F = 5.6) (P < .001 for all). For the glabella, there was a significant effect of time on pliability (F = 32.23), elastic recoil (F = 31.66), and Uv/Ue ratio (F = 10.11) (P < .001 for all). For the lateral orbit, there was a significant effect of time on pliability (F = 15.83, P < .001), elastic recoil (F = 11.43, P < .001), and Uv/Ue ratio (F = 10.60, P = .009).
Conclusions and Relevance
This study provides further evidence that there is an alteration in biomechanical properties of the skin after injection with onabotulinum toxin A. This effect appears to last up to 4 months in the glabella and up to 3 months at other sites. The decrease in the Uv/Ue ratio suggests onabotulinum toxin A injection does not result in an increase in tissue edema suggestive of an inflammatory reaction within the skin. However, it remains unclear whether this is due to an intrinsic property of the medication or another unrecognized mechanism.
Level of Evidence
Human skin has 3 key biomechanical features: strength, pliability (or compliance [the ability to stretch]), and elasticity (or resilience [the ability to recoil]). As one ages, these biomechanical properties change.1 Of these age-related changes, the loss of skin elasticity appears to be the most prominent.1- 8
Physicians often use a variety of methods to attempt to reverse the signs of aging. One such method involves the use of onabotulinum toxin A (Botox; Allergan Inc) injection. Although the primary mechanism of action of onabotulinum toxin A involves a paralysis of the treated muscle that is due to blocking transmitting at the motor end plate, recent evidence suggests that the use of onabotulinum toxin A also results in an alteration in the elasticity and pliability of the skin.9
In this study, we used nonstandardized dosing of onabotulinum toxin A injected into the glabella and lateral orbit. Furthermore, the study only measured the results up to 2 months after treatment; thus, it is unclear whether the results are maintained or return to normal after a short period. Given that onabotulinum toxin A lasts a mean of 3 months,10- 20 one would expect the changes that occur in skin elasticity to return to the original pretreatment levels.
Understanding the natural course that onabotulinum toxin A has on the elasticity of skin may help physicians understand why there appears to be a progressive reduction in wrinkle levels with repeated treatments.21,22 There have been a number of theories as to why progressive wrinkle reduction occurs after repeated injections of onabotulinum toxin A. One theory suggests that it is a result of a learned response, such that patients learn not to use their facial muscles for frowning and thus have less motion contributing to the formation of wrinkles. A second theory suggests that there is a long-term physiologic change in the muscle itself (ie, muscle atrophy).23 Both these theories suggest that the wrinkle formation stops after the motion is inhibited. They do not, however, offer any suggestion as to why the wrinkles fade. On the basis of previous evidence, a third theory suggests that onabotulinum toxin A may have a direct effect on the skin at a histologic level.9,21 However, it remains controversial whether onabotulinum toxin A results in a direct effect on skin cells or whether the alteration in skin biomechanical properties is a result of local inflammation. A study by Dobrev24 noted that in skin that was exposed to UV light radiation, the radiation induced inflammation and caused specific changes in the biomechanical properties of the skin. When a tension force is applied to human skin, there are 2 primary mechanisms that resist the force. Initially, the elastic component of the skin resists the stretch (Figure 1). Once the skin’s elastic resistance (Ue) is at its maximum, a continued force on the skin results in the further skin deformation as a result of the viscoelastic component of resistance (Uv). The Uv of skin expansion is due to displacement of interstitial fluid through the fibrous network of the skin. The results of the study by Dobrev24 noted that when inflammation was present, there was an increase in edema within the skin, resulting in a relative increase in the Uv of skin distention. Of interest, Dobrev24 noted that if the inflammatory response was blocked, there was no increase in the Uv of skin expansion. If the application of onabotulinum toxin A results in an inflammatory change in the skin, it is likely that the Uv will increase relative to the Ue. However, if onabotulinum toxin A is resulting in a change in collagen and elastin levels, we hypothesize that a reduction in Uv and an increase in the Ue would occur.
To gain a further understanding of the effect of onabotulinum toxin A on the skin, a study assessing the effect of the medication may help identify potential intrinsic changes that occur in facial skin after onabotulinum toxin A treatment. Therefore, the purpose of this study is to test the hypothesis that onabotulinum toxin A causes an initial increase followed by decrease in skin elasticity and pliability along with a reduction in the relative contribution of the Uv of skin, reaching a maximum effect at 3 months.
All patients presenting to a facial plastic surgery clinic from October 1, 2012, through June 31, 2013, for an injection of onabotulinum toxin A over the glabella, forehead, and lateral orbit were approached for enrollment. Patients had to have a minimum of mild wrinkle levels at the glabella and lateral orbit according to the Facial Wrinkle Scale.25 This clinic consists of 2 otolaryngologists (J.P.B. and D.E.) practicing in facial plastic surgery. The clinic is associated with the Department of Otolaryngology–Head and Neck Surgery at the University of Toronto, Toronto, Ontario, Canada. The clinic performs 1000 to 1200 onabotulinum toxin A injections a year. All research followed the Declaration of Helsinki, and participants signed a written informed consent form approved by the University of Toronto Research Ethics Board.
Patients were excluded if they met any of the following criteria: pregnant or breastfeeding, a medical comorbidity resulting in a contraindication to onabotulinum toxin A, scars or other anomalies over the measurement sites, any dermatologic conditions underlying the measurement site, any history of onabotulinum toxin A injection, and concomitant or planned skin tightening (both laser or surgical) or any other facial laser procedures during the treatment period or within 2 months before the initial treatment. The use of tanning beds or skin care products was not considered exclusion criteria in this study.
A Cutometer MPA 580 skin elasticity meter along with the accompanying software (Cutometer MPA Q) was used for all patients (Courage & Khazaka Electronic, Cologne, Germany). The Cutometer MPA 580 is a device that is often used to provide an objective measure of in vivo skin biomechanical properties.7,8,26- 30 The Cutometer consists of a probe with a 4-mm round opening that is placed on the skin. The skin contact portion of the probe contains a spring-mounted component that is designed to buffer small variations in force and thus prevent fluctuations in the force applied to the skin.31 The probe is connected to the primary Cutometer hardware by a tube that transfers the suction generated by the primary hardware to the 4-mm probe opening. This suction then deforms the skin, pulling it into the probe. Within the probe, a light measures the deformation of the skin over time as well as the relaxation of the skin after the suction is removed. The Cutometer was calibrated before testing according to company guidelines. Cleaning was performed on the probe daily according to guidelines by the company. A specially designed double-sided adhesive tape was used for all patients during testing to ensure adequate contact between the skin and the probe.
Settings were the same as those of a previously reported study.9 All patients were tested using the time-strain mode (mode 1). For each trial, a 3-second active-suction period followed immediately by a 3-second suction-off relaxation period was used with a setting of 400 millibar (40 kPa) of suction. This 6-second on-off period will be referred to as 1 cycle in this article. One cycle was performed for each trial.
Data were collected at 5 periods: before treatment and 2 weeks, 2 months, 3 months, and 4 months after treatment. The experimental protocol was the same for each period. The same investigator (J.P.B.) collected data for all periods.31 This investigator was masked to the results of the testing until all data collection was completed.
All testing was performed in a humidity- and temperature-controlled environment. The temperature was maintained at 23°C with a relative humidity of 35%. All participants were required to remain in the environment for a minimum of 15 minutes before testing. Participants’ skin was cleaned with isopropyl alcohol solution to remove any residual desquamating skin, moisturizing cream, or cosmetics before testing. The skin was then allowed to dry for 5 minutes.
After this 5-minute rest, we identified 3 testing sites on each patient’s face and 1 on the forearm (control site). One side of the face was chosen for testing. The tested side was determined randomly by a coin toss. The same side was tested for all periods.
Testing site 1 was at the glabella, defined as the point half way between the medial brows above the nasion. Site 2 was the supraorbital point, defined as the point corresponding to a vertical line dropped from the lateral limbus of the eye at a right angle to the Frankfort horizontal plane 1.5 cm above the orbital rim above the brow. This site was chosen as the approximate point at which the corrugator supercilious inserts into the skin.32 Site 3 was the lateral orbit defined as 1 cm lateral to the orbital rim at the midpupillary line. All testing was performed with the patient at repose.
The forearm skin was used as a control. For this site, a distance of 10 cm was measured from the proximal volar wrist crease in the midaspect of the patient’s right forearm. Each site was tested 3 times, and the median value was used for the final data analysis.
The procedure for each measurement was as follows. Once the Cutometer software program was opened on a computer, the investigator selected “measurement” from the main program screen. At this point, the Cutometer MPA 580 increased the internal suction pressure to 400 millibar (40 kPa). The double-sided adhesive tape was then secured to the probe head. Once completed, the investigator placed the probe on the measurement site at the marked location. The probe was then secured at a 90° angle to the skin. At this point, the software was instructed to perform 1 measurement cycle. Once completed, the probe was removed from the patients’ skin, and a 15-second rest was administered to limit the effect of elastic hysteresis. We attempted to maintain light skin pressure for all testing.31,33 Testing was conducted for each trial and skin location. The double-sided tape was replaced every 3 trials.
Onabotulinum toxin A was used for all patients. Each 200-U vial of onabotulinum toxin A was mixed with 2 mL of sodium chloride, 0.9%, with preservatives. All patients received a dose that followed guidelines proposed by Carruthers et al.34 Given that patients were naive to the toxin, a low dose was used for all sites. Female patients received a total dose of 9 U (3 injection points) for each lateral orbit. A total dose of 20 U was used for the glabella (5 injection points). A 30-gauge needle was used for all injections. A maximum of 3 injections was performed with each needle. One injection was performed over the procerus, 3 injections were performed along the corrugator bilaterally, and 3 injections were performed along the orbicularis bilaterally as described by Carruthers et al.35 No additional injections were performed while the patient was enrolled in the study.
There were 3 primary outcome measures collected for this study (Figure 1): pliability, an absolute parameter reflecting the distance the skin is stretched during the 3-second suction period; overall (gross) elasticity, a relative parameter representing the ratio between the elastic recoil after 3 seconds of suction release divided by the pliability1,36; and overall elasticity (or elastic recoil), expressed as the percentage the skin returned to baseline position after 3 seconds of no suction.
When the skin is being stretched, there are 2 components resisting lengthening of the skin: Ue and Uv (Figure 1). The Uv/Ue ratio represents the 2 components that resist the stretching of the skin during the 3-second suction period.
Demographic data, including age, sex, Fitzpatrick skin type, and Facial Wrinkle Scale score, were recorded for each patient. Both primary outcomes were tested to determine whether they differ significantly from a normal distribution using the Anderson-Darling test.37 Any distribution with P < .05 was considered significantly different than a normal distribution, and thus transformations were attempted using a Johnson transformation.37
For each primary outcome measure, a general linear model analysis of variance (ANOVA) with repeated measures was used to determine whether there was a statistically significant difference between times. Tukey post hoc testing was used to assess individual differences within the ANOVA. Data were analyzed using Minitab software, version 15 (Minitab Inc). Because 3 primary outcome measures were collected, a Bonferroni adjustment was performed; thus, statistical significance was set at P < .017.
A sample size calculation for a 1-way ANOVA with 4 levels was calculated. Assuming P = .025 and a mean (SD) difference between groups of 10% (10%) at a power of 80%, a minimum sample size of 28 patients was required.9 The SD of 10% was estimated by pilot testing and previously published research.8,9
Forty-eight patients from a private cosmetic surgery clinic enrolled in the study. Forty-three patients returned for follow-up. The mean (SD) age of patients was 55.2 (11.3) years, and all patients were female. The median Fitzpatrick skin type was 3. The median Facial Wrinkle Scale score was 2 for the glabella and 2 for the lateral orbit. Pliability (P = .47) and elastic recoil (P = .21) data were not significantly different from a normal distribution.
For the supraorbit, there was a significant effect of time on pliability (F = 20.50), elastic recoil (F = 6.92), and Uv/Ue ratio (F = 5.60) (P < .001 for all). For the glabella, there was a significant effect of time on pliability (F = 32.23), elastic recoil (F = 31.66), and Uv/Ue ratio (F = 10.11) (P < .001 for all). For the lateral orbit, there was a significant effect of time on pliability (F = 15.83, P < .001), elastic recoil (F = 11.43, P < .001), and Uv/Ue ratio (F = 10.60, P = .008). Age was not significant as a covariate in any of the models (P = .74 for lateral orbit, P = .62 for supraorbit, and P = .52 for glabella).
For pliability (Figure 2), post hoc testing indicated no significant change between baseline and 2 weeks for any site except the glabella (P = .02). There was, however, a significant increase for all 3 sites between baseline and 2 months (P < .001) and 3 months (P < .001). There was no significant change at 4 months for any site (P = .14 for lateral orbit, P = .79 for supraorbit, and P = .41 for glabella) (Table 1).
For elastic recoil (Figure 3), there was no significant change between baseline and 2 weeks for the lateral orbit (P = .98) and the supraorbit (P = .97), whereas there was a significant effect at the glabella (P = .01). For all sites, there was a significant increase in elastic recoil between baseline and 2 months (P < .001). At 3 months, only the glabella reached significance (P < .001), whereas the lateral orbit (P = .40) and the supraorbit did not (P = .29). At 4 months, only the glabella maintained a significant increase in elastic recoil (Table 1).
When assessing the Uv/Ue ratio (Figure 4), there was no significant difference between baseline and 2 weeks for any site (P = .95 for lateral orbit, P = .63 for supraorbit, and P = .78 for glabella). For the glabella, there was a significant reduction in the ratio compared with baseline at 2 months (P < .001) and 3 months (P = .004) but not at 4 months (P = .90). For the supraorbit, there was a significant reduction in the ratio at 2 months (P < .001), but the ratio at 3 months (P = .03) and 5 months (P = .90) did not reach our definition of significance. For the lateral orbit, only the 2-month period (P < .001) revealed a significant reduction compared with baseline (Table 2). The forearm (the control site) did not have any significant change over time for pliability (P = .46), elastic recoil (P = .32), or Uv/Ue ratio (P = .21).
The results of this study confirm that the injection of onabotulinum toxin A in facial skin results in an increase in pliability and elastic recoil along with a reduction in the Uv/Ue ratio for skin distention. By 4 months, the effect returned to baseline.
A previous study1 assessing skin biomechanical changes using the Cutometer have noted that changes in the skin that occur with aging are the opposite of the changes that occur after an injection of onabotulinum toxin A. Aging38 and UV radiation exposure2,3,39 lead to an increase in elastase activity, resulting in a breakdown of elastin fibers and thus a reduction in skin elasticity.40 From a biomechanical point of view, the application of UV light radiation results in a significant reduction in elasticity and an increase in the Uv and thus an increase in the Uv/Ue ratio. The results of this study indicate that treatment with onabotulinum toxin A results in the opposite of these changes.
Of interest, this study also confirmed the contribution of Uv and Ue to skin stretching change over time after injection of onabotulinum toxin A. One potential hypothesis is that the dermal network of collagen may become more organized during the period when the onabotulinum toxin A is affecting the skin as evident in the Uv/Ue data. As noted in a previous study on fibroblasts,41 the application of onabotulinum toxin A results in increased collagen, elastin, and procollagen levels. As more collagen and elastin is produced, the relative amount of Ue (compared with Uv) increases. These changes produce skin that has the characteristic features consistent with youthful skin.6 After 2 to 3 months, these changes begin to wear off and skin begins to return to its normal characteristics for that individual. This effect corresponds to the mean duration of effect of the medication.
These changes mimic those that one would expect if a change to more youthful skin, as measured by the Cutometer, were to occur. Specifically, as skin ages, the elasticity (elastic recoil) reduces from a mean of 70% at the age of 20 years to near 50% at the age of 70 years.1 Similarly, the pliability of the skin also decreases over time, however, with a lower correlation with age than elastic recoil.1
Nevertheless, it still remains unclear how onabotulinum toxin A results in these skin changes. A study by Oh et al42 assessed the effect of onabotulinum toxin A on human fibroblasts, noting that at 36 and 48 hours there was a significant increase in procollagen, Col 1A1, and Col 1A2 and a reduction in matrix metallopeptidase 9 in a dose-dependent manner compared with controls.
The effects of onabotulinum toxin A on the skin are similar to that reported when using radiofrequency skin tightening. Studies assessing radiofrequency skin tightening have noted multiple potential mechanisms of action. One study43 noted that thermal stimulation of the dermis results in a microinflammatory stimulation of fibroblasts, thus producing new collagen and new elastin, which possibly increases skin tightness. Although some basic evidence exists on the effect of onabotulinum toxin A on fibroblasts, it remains unclear whether the skin-tightening effect is due to the medication effect or a local inflammatory response to the injection.
However, as noted in the study by Dobrev,24 an inflammatory response in the skin results in a reduction in elasticity and an increase in the Uv, the opposite of what was observed in this current study. If onabotulinum toxin A were causing a local inflammatory response, one would expect a similar pattern in the Uv/Ue ratio; however, this is not the case. Nevertheless, the results of the current study provide evidence supporting the notion that the medication is having a direct effect on skin, yet the actual mechanism remains elusive. Of interest, the finding that the effect on the skin biomechanics mimics the duration of action of the product on motor end plates provides some evidence that there may be more than a simple inflammatory response at play.
Although this study provides evidence supporting our hypothesis, there are limitations. This study was not double-blinded, thus reducing the level of evidence. However, the use of an objective measuring device likely limits some of this bias. It also remains unclear whether repeated injections result in larger, more lasting changes in skin. Future studies are required to determine whether these changes continue to occur over time. In addition to this, there was no histologic assessment performed on skin tissue; thus, it is not possible to indicate with confidence that the biomechanical changes are a direct result of collagen and/or elastin changes in the skin.
This study found an increase in skin pliability and elasticity with a corresponding reduction in the Uv after treatment of facial wrinkling with onabotulinum toxin A. The changes occurring in patients’ skin appear to be the opposite of those associated with the aging process1 and UV radiation exposure and inflammation.24 This study also suggests that the duration of effect of these changes mimics the duration of effect of the medication. Future studies are required to determine and quantify the histologic changes that are occurring.
Accepted for Publication: March 18, 2015.
Corresponding Author: James P. Bonaparte, MD, MSc, FRCSC, Department of Otolaryngology–Head and Neck Surgery, University of Ottawa, 1919 Riverside Dr, Ste 308, Ottawa, ON K1S2M3, Canada (Drjames.email@example.com).
Published Online: May 21, 2015. doi:10.1001/jamafacial.2015.0376.
Author Contributions: Drs Bonaparte and Ellis 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.
Acquisition, analysis, or interpretation of data: Bonaparte.
Drafting of the manuscript: Bonaparte.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Bonaparte.
Obtained funding: Ellis.
Study supervision: Ellis.
Conflict of Interest Disclosures: Dr Ellis reported receiving unrestricted funding, speaking at meetings, and serving on the advisory board for Allergan Canada. Dr Bonaparte reported receiving an unrestricted educational research grant from Allergan Canada. No other disclosures were reported.
Additional Information: This study was awarded the John Orlando Roe Award at the 2014 Fall Meeting of the American Academy of Facial Plastic and Reconstructive Surgery; September 18, 2014; Orlando, Florida.
Additional Contributions: Debra Webber, BA, assisted with manuscript preparation. She did not receive compensation for her contribution.