Images of an identical mature scar were generated in linear and zigzag configurations using computer software. These scars were placed on the forehead (flat surface), cheek (convex surface), and temple (concave surface) of 4 volunteers (a white man and woman older than 50 years and a white man and woman younger than 50 years).
The side-by-side scar comparison and 10-point Patient and Observer Scar Assessment Scale are shown.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Ratnarathorn M, Petukhova TA, Armstrong AW, Wang AS, King TH, Eisen DB. Perceptions of Aesthetic Outcome of Linear vs Multiple Z-Plasty Scars in a National Survey. JAMA Facial Plast Surg. 2016;18(4):263–267. doi:10.1001/jamafacial.2016.0107
The process of Z-plasty scar revision breaks up a linear scar into multiple parts with the purpose of camouflage and improvement of the cosmetic appearance of surgical scars. Although this postulation guides the practices of many reconstructive surgeons, few studies support improved aesthetic outcomes.
To compare the perceived cosmetic appearance of linear scars vs zigzag scars by the general public.
Design, Setting, and Participants
A computer-generated image of a mature scar was designed in linear and zigzag configurations and overlaid on the faces of standardized headshots of 4 white individuals. Twelve sets of images of linear vs zigzag scars were arranged in side-by-side comparisons in an Internet-based national survey. Respondents rated each scar on the 10-point Patient and Observer Scar Assessment Scale, where a lower score indicated likeness with normal skin and a higher score, the worst scar imaginable. Data were collected from May 1 through June 30, 2013, and analyzed from July 31 to September 1, 2013.
Main Outcomes and Measures
Aesthetic rating of scars by the survey respondents.
Eight hundred seventy-six participants responded to the survey (24.5% response rate); of these, 810 completed the survey (379 men [46.1%] and 443 women [53.9%]; 148 [18.0%] were 18 to 29 years, 171 [20.8%] were 30 to 44 years, 290 [35.3%] were 45 to 60 years, and 213 [25.9%] were older than 60 years). Significantly lower scores and better perceived cosmetic outcomes were found for linear scars compared with zigzag scars in every assessed group of images on the Patient and Observer Scar Assessment Scale (mean [SD] scores, 2.9 [1.6] vs 4.5 [2.2], respectively; P < .001).
Conclusions and Relevance
The lay public has a significantly better perception of the appearance of linear scars compared with zigzag scars in 3 facial locations (temple, cheek, and forehead) of white patients in various age groups.
Level of Evidence
One of the most important outcomes in cutaneous surgery from a cosmetic perspective is public perception of the resulting scar. Z-plasty scar revision was initially described as a technique to break up the linear scar into smaller, more camouflageable segments.1-3 This postulation guides the choice of many surgical closures and scar revisions today, but only a few studies4,5 have evaluated cosmetic outcomes for this technique. A split-scar study of linear vs zigzag incision in 6 individuals undergoing hysterectomy revealed no significant difference in the color and width of the final scar.4 A randomized trial of temporoparietal facial flaps in 30 patients revealed that a zigzag incision design resulted in a significantly more conspicuous scar than a linear approach in patients younger than 65 years.5
No studies in the published literature, to our knowledge, have compared the public perception of the cosmetic appearance of linear vs zigzag scars. Thus, we conducted a survey study to poll the lay public on their assessment of such scars by presenting side-by-side comparisons of identical computer-generated images of linear or zigzag scars on standardized headshots of white individuals of different age groups.
We designed a prospective, Internet-based survey study targeting the lay public. Data were collected from May 1 through June 30, 2013. Our a priori recruitment goal was 580 survey respondents, assuming α = .05, power of 90%, and effect size of 0.15. We chose a small effect size because both scars were designed to represent good outcomes. The respondents were unaware of the hypothesis to be tested or the origin or authors of the survey. Ethical consent for this study was obtained from the institutional review board of the University of California, Davis, and the trial was registered at clinicaltrials.gov.6 The University of California, Davis institutional review board ruled this study to be exempt from requiring informed consent.
To provide standardized comparisons of the 2 different scar types, a trained software engineer used photo-editing software (Adobe Photoshop, Adobe Systems) to redesign a model mature scar from a prior surgery into a linear or a zigzag configuration and placed each scar on the headshots of 4 volunteers. The dimensions of the linear scar were 1 mm in width and 2.5 cm in length, whereas those of the zigzag scar were 1 mm in width and 4.38 cm in length, accounting for a gain of 75% in length with a 60° angle between segments. Although scars smaller than 1 mm would be considered ideal for surgical outcomes, we opted for a less-than-perfect scar because an excellent outcome might be difficult for survey respondents to visualize. The scars were placed on 3 anatomic locations of the face, including the forehead (flat surface), cheek (convex surface), and temple (concave surface). The volunteers included a white man and woman older than 50 years and a white man and woman younger than 50 years (Figure 1).
We selected white models because they represent most of our patients undergoing cutaneous surgery. We included both sexes to increase external validity and 2 different age groups to account for possible differences in scar perception based on age. Representative flat, concave, and convex surfaces were chosen to gauge whether anatomic curvatures contribute to the perception of scars. Where possible, the zigzag or linear scars were aligned to conceal them along rhytides or oriented along relaxed skin tension lines with the exception of the forehead, where a vertical scar is typically desired to avoid eyebrow elevation.
The primary outcome, the aesthetic rating of linear vs zigzag scars, was assessed using a 10-point scale adapted from the Patient and Observer Scar Assessment Scale,7 with a score of 1 defined as normal skin and 10 corresponding to the worst scar imaginable. Paired images of linear and zigzag scars were generated on each model for each of the designated locations, yielding a total of 12 comparisons (Figure 2). Basic demographic data were obtained from survey respondents, including race, sex, age, educational level, household income, employment status, and US region of domicile.
The survey was distributed to US participants through an online email polling service (SurveyMonkey). This service recruits respondents from the more than 30 million people who use the service each month. Respondents are rewarded with noncash incentives, which include charitable donations and sweepstakes entries. Routine benchmark surveys are administered to ascertain that respondents are representative of the US population as a whole. Image sequence and question order were randomized to minimize order bias.
Data were analyzed from July 31 to September 1, 2013. The data were first summarized descriptively. To assess the differences between linear vs zigzag scar outcomes, we compared the mean Patient and Observer Scar Assessment Scale scores by the location of the scar, model patient sex, and model patient age. Pairwise comparisons were applied to evaluate the primary outcome, and adjustments were performed to account for multiple responses provided by each respondent. With each closure technique (linear or multiple Z-plasty), we assessed whether differences in scar outcomes existed depending on convexity or concavity of the scar location (temple vs cheek) and other comparisons among the scar locations using adjusted pairwise comparisons. In addition, for each closure technique, comparisons between model patient sexes and model patient age groups were also performed using pairwise comparisons. Finally, multivariable regression models were constructed to determine the effect of respondent demographic data (age, sex, educational level, income level, and region) on scar outcome assessment. Respondent regions of domicile were grouped as Pacific, Mountain, Central, and Eastern. All analyses were performed using STATA software (version 9.3; StataCorp).
The survey response rate was 24.5% (876 of 3575 individuals); 810 of 876 respondents (92.5%) completed the survey. Three hundred seventy-nine survey respondents (46.1%) were male and 443 (53.9%) were female; 148 (18.0%) were 18 to 29 years, 171 (20.8%) were 30 to 44 years, 290 (35.3%) were 45 to 60 years, and 213 (25.9%) were older than 60 years. Respondents were from all geographic regions of the United States (Table 1). The mean time to complete the instrument was 4 minutes 37 seconds. Multivariable regression models revealed that respondent age, sex, educational level, income level, and region had no significant effect on the outcome assessment for the linear or the zigzag scar assessments.
Patient and Observer Scar Assessment Scale ratings were statistically significantly lower for linear scars than zigzag scars for every assessment (Table 2). The primary outcome—the difference between mean scores of linear and zigzag scars for all models—was also highly significant (mean [SD] scores, 2.9 [1.6] vs 4.5 [2.2], respectively; P < .001).
In a national survey administered to the lay public, most of the respondents rated the aesthetic appearance of a linear scar more favorably than that of an equivalent zigzag scar. This finding was consistent whether the scar was located on the temple, cheek, or forehead. In addition, linear scars were favored over zigzag ones on all 4 white models of different sexes and ages. The preference for linear scars did not vary by respondent sex, age, educational level, income, or geographic region of residence in the United States.
These findings run contrary to the current dogma on scar perception. Because the study did not investigate rationale for the respondents’ choices, explanations for the results are purely speculative. One explanation might be that an appropriately sized Z-plasty revision of a linear scar is nearly 75% longer when using a traditional 60° angle. Another possibility might be that the zigzag pattern creates a scar with a regular repetitive pattern and, therefore, may appear less natural than a linear scar to some observers. Thus, the increased scar length and unnatural appearance of the zigzag scar may have adversely affected its aesthetic rating.
Prior studies on this topic included a very small number of respondents. Surgical linear or zigzag incisions were placed and resultant scars compared on the abdomen of 6 women4 and temporoparietal scalp of 27 randomized patients.5 Neither of the studies found significant improvement of scar cosmesis with a zigzag approach.
A strength of the present study is the large number of respondents representing all 4 geographic regions of the continental United States. The nature of an Internet-based investigation allows for a much larger number of assessments than is possible within the confines of a randomized clinical trial. Advantages of the virtual approach include that the primary outcome is achieved without the risks of a surgical intervention and that the 2 scar configurations could be compared without the typical postoperative confounders such as variations in scar width, erythema, pigmentation, or topographic abnormalities.
Although our survey included a large number of respondents, a limitation of the study is that the population did not include individuals without access to the Internet, who did not speak English, or who were illiterate. The depiction of relatively good linear and zigzag scars is also a limitation because we did not account for potential adverse results from surgical treatment. Complications such as dehiscence, scar spread, topographic irregularities, or even unintended depression of free margins as a result of the Z-plasty may have occurred in the setting of a randomized clinical trial. In addition, the virtual 2-dimensional photographic image cannot account for potential scar irregularities that may affect cosmesis in a 3-dimensional real-life situation. Furthermore, our study included a limited number of anatomic locations and model subjects, thus reducing its applicability to only a few facial sites. We also acknowledge that surgeons perform Z-plasties for reasons other than cosmetic concerns, such as the release of contractures and webs of scars. Clearly, our study does not address those purposes.
The present study highlights the need for a large, multicenter, randomized clinical trial to best assess the cosmetic outcomes of linear vs multiple Z-plasty closure techniques. Until that time, traditional dogma related to aesthetic preferences regarding Z-plasty scars should be interpreted with caution.
Our findings suggest that a linear scar on the temple, forehead, or cheek is more aesthetically acceptable to the lay public than an appropriately sized zigzag scar on the same locations. Until randomized clinical data become available on this topic, surgeons should consider carefully whether the extra time and effort involved in performing Z-plasties provide additional cosmetic benefit outside of the need for functional improvements, such as contracture, web revision, or free-margin corrections.
Corresponding Author: Daniel B. Eisen, MD, Department of Dermatology, University of California, Davis, 3301 C St, Ste 1400, Sacramento, CA 95816 (firstname.lastname@example.org).
Accepted for Publication: November 24, 2015.
Published Online: April 7, 2016. doi:10.1001/jamafacial.2016.0107.
Author Contributions: Dr Eisen had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Ratnarathorn, Armstrong, King, Eisen.
Acquisition, analysis, or interpretation of data: Petukhova, Armstrong, Wang, Eisen.
Drafting of the manuscript: King, Eisen.
Critical revision of the manuscript for important intellectual content: Ratnarathorn, Petukhova, Armstrong, Wang, Eisen.
Administrative, technical, or material support: Armstrong, King.
Study supervision: Eisen.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This paper was presented at the 2013 Annual Meeting of the American Society for Dermatologic Surgery; October 4, 2013; Chicago, Illinois.
Additional Contributions: We thank the volunteers for granting permission to publish this information.
Create a personal account or sign in to: