Wound healing, a complex and dynamic interactive process, is divided into 3 overlapping phases: inflammation, tissue formation, and tissue remodeling [
Among preventive treatments available, onion extract-based topical gel has been marketed as a product to improve the appearance and texture of surgical scars [
The study was conducted at the Division of Dermatology, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand, with the approval from of the university Institutional Review Board (IRB). The research protocol conformed to the guidelines of the Helsinki Declaration, and informed consent was obtained from subjects prior to enrollment.
Woman who underwent cesarean section with Pfannenstiel’s incision were recruited from the Department of Obstetric and Gynecology. Eligible subjects were ones that have undergone cesarean section for the first time using absorbable suture materials and were able to follow study protocol. They were excluded if they had history of onion allergy or developed surgical complications.
The study was designed as a double-blinded, randomized, split-scar study. Using a randomization table, each side of scar was randomly assigned into receiving either a vehicle-based gel or an onion extract gel containing 12%
The outcomes were assessed by both objective and subjective evaluation at baseline and 4th and 12th week after starting the treatment. Photographic documentation was performed from the front side of the scar under the same conditions at baseline and each subsequent visit.
For objective evaluation, colorimeter (ChromaMeter CR-231, Minolta Corporation, Osaka, Japan)was used to assess scar redness. We used the Commission Internationale de l’ Eclairage L*a*b* color system, which allowed a color to be quantified on three axes: white-black (L*), red-green (a*), and yellow-blue (b*). We used the a* parameter with values from +60 to −60 (green) to measure subject’s skin redness index (SRI) [
Physician and subject evaluation were performed at baseline and on each subsequent visit. Two blinded investigators rated the scars independently on each side using 2 parameters: pliability and height. Grading of pliability ranged from 0 to 5: 0: normal pliability; 1: supple, flexible with minimal resistance; 2: yielding (giving way to pressure, offering moderate resistance but did not behave as a solid mass of scar); 3: firm (solid, inflexible unit, not easily moved, resistant to manual pressure); 4: banding (rope-like tissue that blanched with extension of scar, did not limit range of motion(ROM)); 5: contracture (permanent shortening of the scar producing deformity or distortion, limited ROM). Scar height was graded as 0 (flat), 1(<2 mm), 2 (2 to 5 mm), and 3(>5 mm). All participants scored the visual analogue scale ranging from 0 (absent) to 3(severe) to evaluate each side for pain, itching, discomfort, tightness, and hardness. This was added to give a total summation of 15. Subjects were also asked to rate the overall cosmetic improvement of each side of the scar using scores −1 (worse), 0 (no improvement), 1 (minimal improvement), 2 (moderate improvement), and 3 (marked improvement).
Any adverse effects that occurred after applying the products were recorded at each visit.
Statistical analysis was performed by using computer software (STATA/SE version 11.2, STATA Corp, College Station, TX). Categorical variables (i.e., gender, Fitzpatrick’s skin type, and gravidity) were expressed as percentages and continuous variables (i.e., age, scar redness, scar pliability, scar height, scar symptoms, and overall cosmetic appearance) were expressed as mean. To compare the variables between the control group and the treatment group, a paired
Of the 26 subjects who were assessed for eligibility, 6 were excluded (3 declined to participate and 3 did not meet the inclusion criteria). Twenty female participants who met the eligibility criteria were enrolled. There were sixteen subjects who completed the protocol. Four subjects did not complete the study due to incomplete follow-up (Figure
The flow chart showing number of eligible, refused to participate, recruited, randomized, and completed subjects.
All of the participants were Thai women with Fitzpatrick’s skin types varying from type 3 to 5, the majority were skin type 4 (75%). Subjects’ age ranged from 19 to 43 years, with the median age of 31 years. Nine subjects (56.25%) were nulliparous and 7 subjects (43.75%) were multiparous (Table
Demographic data of the subjects.
Demographic characteristic | Number of subjects (%) |
---|---|
Gender | |
Female | 16 (100%) |
Age | |
19–23 years | 1 (6.25%) |
24–28 years | 3 (18.75%) |
29–33 years | 6 (37.50%) |
34–38 years | 4 (25%) |
39–43 years | 2 (12.50%) |
Gravida | |
Nulliparous | 9 (56.25%) |
Multiparous | 7 (43.75%) |
Fitzpatrick’s skin type | |
Type 3 | 1 (6.25%) |
Type 4 | 12 (75%) |
Type 5 | 3 (18.75%) |
Seven subjects (43.75%) had a lower SRI on the onion extract side compared to control side at the 4th week and 5 subjects (31.25%) had a lower SRI on the onion extract side at the 12th week. The mean SRI decreased for both control and treatment sides at weeks 4 and 12. On the control side, the mean SRI changed from
There was no difference in mean scar height between the control side and the treatment side at baseline. Six patients (37.50%) had a lower scar height on the onion extract side compared to the control side at the 4th week and 9 subjects (56.25%) had a lower scar height on the treatment side at the 12th week. A statistically significant difference in mean scar height between the two sides was found at the 4th week (
Photograph of a 32-year-old female subject at the 12th week follow-up. Onion extract and placebo were applied on the right and left side, respectively. Hypertrophic scar was visibly more elevated in the left side.
This figure demonstrates mean scar height on the placebo compared to the treatment side. A statistically significant difference favoring the treatment side was found at the 4th and 12th week follow-up. Red and blue line represents placebo and treatment side, respectively.
There was no difference in mean scar pliability score between the two sides at baseline. Three patients (18.75%) had a lower scar pliability score on the onion extract side compared to the control side at the 4th week and 7 subjects (43.75%) had a lower scar pliability score on the onion extract side at the 12th week. However, this was not statistically significant (4th week:
At baseline, there was no difference in mean scar symptoms between the control side and the treatment side (
This figure compares mean scar symptoms score on the placebo and the treatment side. A statistically significant difference was found at the 4th and 12th week follow-up. Red and blue line represents placebo and treatment side, respectively.
At the 4th week, all 16 control sides and 15 of 16 treatment sides were rated as no improvement (overall cosmetic improvement score = 0). One of 16 treatment sides was rated as worse (overall cosmetic improvement score = −1). At the 12th week, 15 of 16 control sides and 14 of 16 treatment sides were rated as no improvement; 1 of 16 control sides and 1 of 16 treatment sides were rated as minimal improvement (overall cosmetic improvement score = +1); 1 of 16 treatment sides was rated as moderate improvement (overall cosmetic improvement score = +2). There was no statistically significant difference in overall cosmetic improvement score at any point of evaluation.
Overall, subjects tolerated both onion extract and placebo gel well. No adverse effect was observed.
It is difficult to predict scar development following surgical wound. It can vary from fine asymptomatic scars to hypertrophic scars and keloids leading to pruritus, pain, disfigurement, and psychological stress [
Various treatment options exist for treating hypertrophic scars and keloids [
In this study, we selected split-scar design to have optimal internal control. Pfannenstiel’s cesarean section scar was chosen instead of midline scar to control the anatomical variation, knowing that the right and left side of this transverse incision should be more or less similar. Hence, the scar comparison in this study had matching location in the same patient with the identical cutaneous injury. Furthermore, the internal control design of this study eliminated confounding factor of differences in suture material, diverse surgical technique, and experience of surgeon. In addition, all of our subjects were Asian women with skin type 3 to 5. By choosing a group inherently of high risk of developing hypertrophic scars, we hoped to enhance a significant result.
The result of our study showed a significant difference in mean scar height and mean scar symptoms between control side and the treatment side at weeks 4 and 12. Although there were no statistically significant difference in mean scar pliability scores at any point of evaluation, a nearly significant difference was found at the 12th week (
Jackson and Shelton [
According to a randomized, double-blinded, split-scar study by Chung et al. [
In general, all subjects tolerated the products well without any reports of adverse effects. Unlike the study by Jackson and Shelton, irritation was reported in 33.33% of subjects in the onion extract group [
The limitations in this study were the small sample size and the lack of long-term follow-up, since keloids may not develop until years after the event [
The use of 12% onion extract gel on cesarean section scars three times daily during early postoperative period reduced height and scar symptoms without side effects. There was tendency towards improvement of scar pliability. Scar redness and overall cosmetic results did not improve. It can be used as adjunctive treatment to prevent hypertrophic scar from surgical wounds.
The authors thank Drs. Chulaluk Komoltri and Pinyo Rattanaumpawan for their kind assistance and valuable statistical analysis. 12%