Botulinum Toxin to Improve Facial Wound Healing:
A Prospective, Blinded, Placebo-Controlled Study

HOLGER G. GASSNER, MD; ANTHONY E. BRISSETT, MD; CLARK C. OTLEY, MD; DEREK K. BOAHENE, MD; ANDY J. BOGGUST, MD; AMY L. WEAVER, MS; AND DAVID A. SHERRIS, MD

OBJECTIVE: To test whether botulinum toxin–induced immobilization of facial lacerations enhances wound healing and results in less noticeable scars.

PATIENTS AND METHODS: In this blinded, prospective, randomized clinical trial, patients were randomized from February 1, 2002, until January 1, 2004, to botulinum toxin vs placebo injection into
the musculature adjacent to the wound within 24 hours after wound closure. Blinded assessment of standardized photographs by experienced facial plastic surgeons using a 10-cm visual analog
scale served as the main outcome measure.

RESULTS: Thirty-one patients presenting with traumatic forehead lacerations or undergoing elective excisions of forehead masses were included in the study. The overall median visual analog scale
score for the botulinum toxin–treated group was 8.9 compared with 7.2 for the placebo group ( P=.003), indicating enhanced healing and improved cosmesis of the experimentally immobilized scars.

CONCLUSIONS: Botulinum toxin–induced immobilization of forehead wounds enhances healing and is suggested for use in selected patients to improve the eventual appearance of the scar. Mayo Clin Proc. 2006;81(8):1023-1028

IRB = institutional review board; VAS = visual analog scale

The presence of a facial scar may profoundly affect an individual’s psychological well-being. Surgeons seek to achieve the most aesthetic scar when placing an elective incision or repairing traumatic lacerations. Immobilization is a basic therapeutic principle to aid in the healing of various types of tissue. A major factor that determines the final cosmetic appearance of a cutaneous scar is the tension that acts on the wound edges during the healing phase. Surgical techniques carefully executed to best align incisions with the relaxed skin tension lines are routinely applied in facial plastic surgery. However, such techniques reduce rather than eliminate the muscle tension that acts on the healing wound. Repeated microtrauma, caused by the continuous muscle activity around a skin wound, induces a prolonged inflammatory response and increased metabolic activity during the healing process. Consequently, extracellular deposition of collagen and glycosaminoglycans can intensify and lead to hypertrophic and hyperpigmented scars.1

Since Carruthers and Carruthers2 described the cosmetic use of botulinum toxin in the face, it has been used extensively to treat facial rhytides. Paralysis of facial muscles subjacent to wounds can reliably be achieved by injection of botulinum toxin. This method has been shown to result in cosmetically more favorable scars in a controlled primate model.3 Human case reports also suggest that chemoimmobilization of cutaneous facial wounds results in favorable healing.4,5 To our knowledge, the first human report occurred in 2000,6 and prospective controlled trials to elucidate the effect of chemoimmobilization on human facial lacerations or incisions have not been performed. The current study was designed to test whether botulinum toxin–induced chemoimmobilization of facial lacerations affects the cosmetic appearance of the healed wound.

PATIENTS AND METHODS
This study was designed as a prospective, blinded, randomized, placebo-controlled, single-institution trial. Wounds limited to the forehead were to be studied because preliminary data suggesting efficacy and safety were available for this area. To make the results of the trial applicable to the routine clinical setting, wound care and closure technique were performed within the variations of usual clinical practice.

Unselected patients older than 18 years with traumatic forehead lacerations presenting to the emergency department and patients undergoing elective excision of cutaneous neoplasms of the forehead were enrolled. The following inclusion criteria were applied: (1) not currently pregnant,
no potential for pregnancy, or breastfeeding an infant and (2) no history of radiation therapy or chemotherapy, hematologic disorders, neuromotor disorders, or keloids. Enrolled patients were offered remuneration if they completed the following: (1) presented for follow-up visit and photographic documentation at 1 week and 6 months postoperatively and (2) agreed to report any untoward events or
adverse effects to the study coordinator. The study protocol was approved by the Mayo Foundation Institutional Review Board (IRB). Patients were enrolled from February 1, 2002, until January 1, 2004.

The patients with traumatic forehead lacerations were seen in the emergency department at the Mayo Clinic and were cared for by the services of otorhinolaryngology, plastic surgery, or emergency medicine. These services performed wound closure and instructed patients in postoperative
care according to their respective clinical routine. After wound closure, the patients were invited to enroll in the study and were seen the following day by the study coordinator and informed about the study. If IRB-approved written informed consent was given, photographs were taken and the wound was injected according to a standardized protocol as outlined herein. Size and location of the traumatic lacerations, as well as the use of suture material and closure technique, are outlined in Table 1.

Patients undergoing elective excisions of the forehead for removal of skin neoplasms were informed about the study protocol before the procedure. If IRB-approved written informed consent was given, the excision of the lesion and wound closure were performed by the primary dermatologic or otolaryngologic surgeon of the patient. Size and location of the excisional wounds, as well as the use of suture material and closure technique, are outlined in Table 1.

TABLE 1. Location and Size of the Forehead Wounds in the Study Population*

*BCC = basal cell carcinoma; SCC = squamous cell carcinoma.
†Listed as the greatest length in centimeters as measured intraoperatively. Measurements obtained on intraoperative photographs are marked with a dagger.
‡Closure technique as listed in the operative note and verified with postoperative photographs. Use of suture material (Vicry, Monocryl, PDS, Prolene, or Nylon; Ethicon, Somerville, NJ) as listed in the operative note.

The Mayo Clinic’s Pharmacy Services was notified and prepared the study medication according to a secure randomization sheet. The randomization schedule was generated using a block approach and stratified by type of wound (excisional wounds or lacerations). Encoded vials of study medication were prepared. The placebo vials contained 0.2 mL of preservative-free normal saline; the experimental
vials contained 0.2 mL of preservative-free normal saline with 15 U of botulinum toxin A (Botox, Allergan, Irvine, Calif). One vial was prepared for patients with wounds up to 2 cm in intraoperative length, 2 vials were prepared for wounds larger than 2 to 4 cm in intraoperative length, and 3 vials were prepared for wounds greater than 4 cm in intraoperative length (Table 1). Three milliliters of a solution of 1% lidocaine with epinephrine 1:100,000 (Xylocaine, Astra Zeneca, Wilmington, Del) was added per vial by the injecting physician immediately before injection. The study medication was then injected into the musculature around the wound by a surgeon experienced in botulinum toxin treatment of the face (A.E.B., C.C.O., D.A.S., H.G.G., D.K.B.). The injections were placed into the musculature adjacent to the wound in a diameter of approximately 1 to 3 cm around the wound edges. The lateral supraorbital rim was spared to prevent lid ptosis.

Photographs of both excisional and traumatic wounds were taken by Mayo Clinic’s Section of Patient and Clinical Photography at the closure of the wounds, at 1-week follow-up, and at 6-month follow-up according to the following protocol. A Nikon 660 Digital Camera with 105- mm Nikon Micro Nikkor Lens (Nikon Inc, Melville, NY) and Nikon SB 21 Flash—M 1/4 Power (Nikon Inc) was used. Photographs were taken at a 1:5 ratio, aperture F/20; close-up photographs were taken at a 1:1 ratio, aperture F/45, at 250th-second shutter speed, ASA 200, and 6- megapixel resolution. The photographs were printed by the Division of Media Support Services at Mayo Clinic on photographic paper.

At the 1-week and 6-month visits, the patients were seen and examined by the study coordinator and operating surgeon. Occurrence, duration, and nature of possible adverse effects were documented.

After the study was ended, 2 experienced facial surgeons (Pamela K. Phillips, MD, C.C.O.) were asked to assess the photographs in an independent and blinded fashion. Considering the condition of the wound at closure and at 6 months, the observers were asked to rate the final cosmetic outcome on a 10-cm visual analog scale (VAS) (0 indicating the worst possible outcome and 10 indicating the ideal outcome). The raters assessed the first patient together to establish a reference score for their respective further scores. The photographs of the subsequent patients were assessed independently.

The agreement between the 2 raters’ VAS scores was determined on the basis of calculating the intraclass correlation coefficient and corresponding 95% confidence interval. Using the average of the 2 VAS ratings for each patient, the VAS scores were compared between the 2 treatment groups using the Wilcoxon rank sum test. All calculated P values are 2-sided, and P<.05 is considered statistically significant.

The study was designed to enroll 20 patients per treatment group. Using the data from 2 previously published clinical trials, Quinn et al7 reported that the minimal clinically important difference on the VAS cosmesis scale is 1.5. These authors also reported a pooled SD of 1.4 among VAS scores for lacerations and 2.2 among VAS scores for incisions, yielding effect sizes (mean difference/SD) of 0.7
to 1.1. A sample size of 16 patients per treatment group would provide an 80% chance (statistical power) of detecting an effect size of 1.0, based on a 2-sample t test with a type I error level of 5%. Assuming a 10% noncompliance rate with the follow-up visit and given that the efficiency of the 2-sample t test relative to the Wilcoxon rank sum test is 95%, the sample size was increased to 20 per group.

RESULTS
Of the 42 patients enrolled in the study, 22 received the experimental drug. Of these 22 patients, 16 completed the study, and 6 were excluded for the following reasons: 4 lacked photographs (eg, were seen after hours when photography was not available) and 2 were lost to follow-up. Twenty patients received the control drug. Of these 20 patients, 5 were excluded from the study for the following
reasons: 1 was lost to follow-up, 1 lacked final photographs, 2 received corticosteroid injections at 4 and 4.5 months after poor wound healing, and 1 underwent a brow lift procedure at 4 months, before the 6-month follow-up visit (Figure 1). Age, sex, and medical comorbidities of the patients included in the study are listed in Table 2. Coincidentally, all patients were white.

FIGURE 1. CONSORT flow diagram.

One placebo patient noted mild headaches that persisted past the 6-month follow-up visit. No other complications or adverse effects, such as lid ptosis, occurred.

The concordance between the 2 physician raters who evaluated the 6-month photographs using a VAS was estimated using the intraclass correlation coefficient. The overall concordance in the VAS scores by the 2 physician raters was 0.86 (95% confidence interval, 0.81-0.93), denoting substantial concordance.

The VAS results, based on the average of the 2 physician ratings for each patient, were analyzed. The overall median VAS for the botulinum toxin–treated group was 8.9 compared with 7.2 for the placebo group. On the basis of the Wilcoxon rank sum test, this difference was statistically significant (P=.003). Three patients had been excluded because they received corticosteroid injections or underwent
a brow lift procedure. All 3 of these patients had received the control treatment. Photographs for evaluation of these patients were available with shorter follow-up (4 months). When these patients were included in the statistical analysis, the outcome did not change substantially (median VAS, 8.9 vs 7.1; P=.002).

Figures 2 and 3 are representative examples of experimentally treated and placebo-treated wounds. Figure 2 depicts a botulinum toxin–treated (15 U of botulinum toxin in 3 mL of 1% lidocaine with 1:100,000 epinephrine), wellhealed scar of the lateral forehead. Figure 2 shows a placebo-treated vertical midline forehead wound of the central forehead with notable scarring.

DISCUSSION
Statistically significant and clinically relevant improvement of the cosmetic appearance of facial scars after treatment with botulinum toxin A was previously shown in a primate model. Human case reports with long-term followup also suggest safety and efficacy of this treatment, with no adverse effects observed.4,5 Choi et al8 used botulinum toxin to prevent complications in a series of 11 patients at
high risk of impaired wound healing. These patients underwent blepharoplasty without postoperative complications. Zimbler et al9 observed a longer-lasting effect of laser skin resurfacing in the face when the treated skin was immobilized with botulinum toxin. Tollefson et al10 described botulinum toxin–induced immobilization for cleft lip repair. To our knowledge, the current study is the first randomized, placebo-controlled, prospective trial to investigate the effect of chemoimmobilization on the eventual cosmetic appearance of incisional and traumatic facial wounds.

FIGURE 2. Left, A 44-year-old white woman 1 week after excision of a basal cell carcinoma of the right lateral aspect of the forehead. The longest postoperative excisional size was 2.7 cm. Fifteen units of botulinum toxin was injected, and layered closure was performed with 5-0 Monocryl subcutaneous and 6-0 Nylon simple running sutures. Right, The resulting scar 6 months after botulinum toxin treatment displays good color match and no hypertrophy or inversion.

Outcome was based on the blinded evaluation by 2 experienced dermatologic surgeons of standardized photographs obtained at the 6-month visit. These examiners assessed the photographs of patient 1 (Table 1) together to establish a reference score. Concordance in the evaluation
of the remainder of the patients’ photographs was substantial. The examiners had photographs of the wound after closure in all cases, allowing assessment of the final cosmetic outcome in light of the condition of the original wound and closure technique. As listed in Table 1, closure technique was consistent. All wounds were closed in a layered fashion, with most wounds closed by a dermatologic
surgeon using nonresorbable cutaneous sutures.

Potential limitations of the current study must be considered. The current trial was limited to a single institution. The patient population was heterogeneous, consisting of both traumatic and incisional wounds. However, most wounds were incisional in both the experimental and the control group. Trauma accounted for 4 wounds in the experimental and 3 wounds in the control group. Therefore,
the different wound types were evenly distributed between the treatment groups. The differences in outcome among the remaining incisional wounds remained statistically significant when the traumatic wounds were excluded from the statistical analysis. The number of patients who did not finish the study or who were excluded must also be considered. Eleven of the 42 enrolled patients were not evaluated
in the final analysis: 8 were lost to follow-up or did not have their photographs taken, and 3 were excluded because they underwent interventions that could influence the outcome (corticosteroid injections, brow lift procedure) before 6 months of follow-up were completed. When we included photographs of these patients with shorter follow-up (4-4.5 months) in the analysis, no relevant change in the overall outcome was observed. Other potential confounding factors, such as skin type, were not controlled for. Current nicotine use was rare, as outlined in Table 2.

FIGURE 3. Left, A 67-year-old white woman 1 week after excision of a squamous cell carcinoma of the central forehead. Postoperative excisional size was 4.0 by 6.0 cm. Six milliliters of the placebo medication (1% lidocaine with 1:100,000 epinephrine) was injected, and layered wound closure was performed with 5-0 Monocryl and 6-0 running Nylon suture. Right, The resulting scar at 6 months after placebo treatment shows notable widening of the scar.

Follow-up was 6 months throughout the study group. Clinically, the resulting scars appeared mature. Quinn et al7 showed that the cosmetic appearance of traumatic facial wounds at 3 months is highly predictable of the appearance at 1 year. Therefore, it seems unlikely that the cosmetic appearance of the wounds would have changed to a relevant degree with longer follow-up.

CONCLUSIONS
Considering these limitations, previous data in the literature, and the substantial and statistically significant differences between the treatment groups in the current study, the following conclusions appear justified. First, no adverse effects of the experimental treatment were observed. In light of previous data and the common use of botulinum toxin in the face, the treatment of facial wounds with
botulinum toxin appears to be safe in the hands of an experienced surgeon. Second, forehead wounds immobilized with botulinum toxin are shown to heal with a superior cosmetic appearance at 6 months. We believe that these data justify the use of chemoimmobilization in selected patients concerned with the eventual appearance of a facial wound. Third, as with every new concept in medicine, more and potentially multicentered trials with longterm follow-up are needed to solidify the current data and
to expand the indications of chemoimmobilization to other areas of the face, neck, and body. Different brands and serotypes of botulinum toxin may be studied for use in cutaneous wound healing.

REFERENCES
1. McCarthy JG. Introduction to plastic surgery, In: McCarthy JG, ed. Plastic Surgery. Vol. 1. Philadelphia, Pa: WB Saunders; 1990:43-44.
2. Carruthers A, Carruthers J. Aesthetic indications for botulinum toxin injections [letter]. Plast Reconstr Surg. 1995;95:427-428.
3. Gassner HG, Sherris DA, Otley CC. Treatment of facial wounds with botulinum toxin A improves cosmetic outcome in primates. Plast Reconstr Surg. 2000;105:1948-1953.
4. Sherris DA, Gassner HG. Botulinum toxin to minimize facial scarring. Facial Plast Surg. 2002;18:35-39.
5. Gassner HG, Sherris DA. Chemoimmobilization: improving predictability in the treatment of facial scars. Plast Reconstr Surg. 2003;112:1464-1466.
6. Gassner HG, Sherris DA, Otley CC, Kienstra MA. Pharmacological immobilization of cutaneous wounds results in more favorable scars in primates [abstract]. Laryngorhinootologie. 2000;79(1, suppl):S81.
7. Quinn J, Wells G, Sutcliffe T, et al. Tissue adhesive versus suture wound repair at 1 year: randomized clinical trial correlating early, 3-month, and 1-year cosmetic outcome. Ann Emerg Med. 1998;32:645-649.
8. Choi JC, Lucarelli MJ, Shore JW. Use of botulinum A toxin in patients at risk of wound complications following eyelid reconstruction. Ophthal Plast Reconstr Surg. 1997;13:259-264.
9. Zimbler MS, Holds JB, Kokoska MS, et al. Effect of botulinum toxin pretreatment on laser resurfacing results: a prospective, randomized, blinded trial. Arch Facial Plast Surg. 2001;3:165-169.
10. Tollefson TT, Senders CM, Sykes JM, Byorth PJ. Botulinum toxin to improve results in cleft lip repair. Arch Facial Plast Surg. 2006;8:221-222.