Evidence‐based therapy in hypertrophic scars: An update of a systematic review

Abstract Hypertrophic scars are still a major burden for numerous patients, especially after burns. Many treatment options are available; however, no evidence‐based treatment protocol is available with recommendations mostly emerging from experience or lower quality studies. This review serves to discuss the currently available literature. A systematic review was performed and the databases PubMed and Web of Science were searched for suitable publications. Only original articles in English that dealt with the treatment of hypertrophic scars in living humans were analyzed. Further, studies with a level of evidence lower than 1 as defined by the American Society of Plastic Surgeons were excluded. After duplicate exclusion, 1638 studies were screened. A qualitative assessment yielded 163 articles eligible for evidence grading. Finally nine studies were included. Four of them used intralesional injections, four topical therapeutics and one assessed the efficacy of CO2‐laser. Intralesional triamcinolone + fluorouracil injections, and topical pressure and/or silicone therapy revealed significant improvements in terms of scar height, pliability, and pigmentation. This systematic review showed that still few high‐quality studies exist to evaluate therapeutic means and their mechanisms for hypertrophic scars. Among these, most of them assessed the efficacy of intralesional triamcinolone injections with the same treatment protocol. Intralesional injection appears to be the best option for hypertrophic scar treatment. Future studies should focus on a possible optimization of infiltrative therapies, consistent end‐point evaluations, adequate follow‐up periods, and possibly intraindividual treatments.


| INTRODUCTION
Scar formation is the physiological response to trauma and the following wound healing cascade of human tissues. Hypertrophic scars are pathologically deviating phenomena that may occur upon an intrinsically or extrinsically altered wound healing cascade. Hypertrophic scars occur in about 30% to 50% after surgery or trauma. 1 An even higher prevalence can be seen after burn injuries. 2,3 These scars lead to symptoms like pain, itchiness, or in severe cases restricted mobility due to loss of elasticity and contracture, besides psychological and cosmetic disturbance. Mechanisms causing hypertrophic scar development are still under debate. Most studies implicate that severe inflammation of the wound affect hypertrophic scar formation. 4,5 Inflammation is crucial for wound healing and is needed for an adequate defense against pathogens as well as for clearing the wound area of debris. In general, wound healing is highly dynamic and selflimiting. Dysregulated wound healing, characterized by prolonged or increased inflammation is correlated with an overproduction of immature collagen III in contrast to mature collagen I, resulting in increased tissue fibrosis. [6][7][8][9] One of the key players in tissue fibrosis, hence collagen production, is transforming growth factor beta (TGF-β) with its subtypes TGF-β1 and −2 as profibrotic and TGF-β3 as antifibrotic isoforms. 10 TGF-β is involved in triggering events throughout all phases of wound healing. 11 During inflammation, TGF-β acts as a potent chemoattractant for neutrophils and macrophages, it regulates immune cell function, and also contributes to resolution of inflammation. In the proliferative phase, TGF-β has been reported to promote angiogenesis by stimulating endothelial cell migration, differentiation, and capillary tubule formation. Moreover, fibroblast proliferation, fibroblast trans-differentiation into myofibroblasts, and ECM production, which show abnormal patterns in hypertrophic scars, are mediated by TGF-β. In addition, TGF-β inhibits keratinocyte proliferation and enhances keratinocyte migration, promoting re-epithelialization.
Finally, TGF-β regulates the balance of ECM synthesis and degradation by tightly controlling the production of ECM components and regulating their rate of degradation in the remodeling phase. 11 An imbalance of these and other involved cytokines are at least coresponsible in the formation of pathologic scars. Recent results have further confirmed epidermal Foxn1 as a relevant transcription factor in the expression pattern of TGF-β subtypes. 12 The exact pathomechanism, however, is not comprehensively elucidated yet.
The to date known findings of the influence of TGF-β and abnormal proliferation suggest interference in these pathways with promotion of apoptosis or abnormal cells as a possible treatment strategy. 13,14 These insights in the molecular pathways of tissue fibrosis being highly correlated to the inflammatory response can further be a possible explanation for the much higher prevalence of pathologic scars after thermal trauma, since thermal trauma is accompanied by significant systemic, 15 and local inflammatory responses. 16,17 The complexity and partially still elusive mechanism behind pathologic scar formation might be one reason for the multitude of treatment regimens available. Many of those are based on suggestions, assumptions, and experience and do somehow interfere with tissue fibrosis and the above-mentioned pathways. Silicone gel treatment and intralesional injection of immunomodulatory drugs have recently emerged as the most promising treatment regimens, with high quality, randomized, (placebo-)controlled studies, still being scarce. The exact mechanism of action of silicone sheeting is still inconclusive; increased temperature, 18 increased hydration, 19 polarized electric charge leading to scar shrinking, 20 and others are being discussed. 21,22 The mechanisms of immunomodulatory drugs are studied more intensively. Especially intralesional triamcinolone (TAC), a glucocorticoid suppressing the inflammatory response, and verapamil (a calcium channel antagonist reducing the synthesis of extracellular matrix) injections have been investigated with regard to their ability to alter the TGF-β expression patterns, 23,24 and to positively influence collagen production. 25 5-Fluorouracil (5-FU) as another rising substance in scar treatment is a pyrimidine anaologue modulating the inflammatory response by inhibiting cell growth, and inducing apoptosis and G2 cell-cycle arrest among others. 26,27 Another novel point of action is the hormone angiotensin II: It has been shown to have profibrotic effects, 28 hence an inhibition of angiotensin-converting-enzyme (ACE) represents a reasonable strategy that has been shown effective in animals. 29 Despite recent advances in the understanding of those substances' mechanisms, studies were still not ultimately able to develop a clinically relevant and effective treatment protocol. One reason for this shortcoming is not least the current lack of an ideal (animal) model for hypertrophic scars, 30,31 which is needed for thorough understanding of the pathomechanism and potential working points. 32 Even though advances could be achieved in the red Duroc pig model, the transferability to human scarring remains uncertain. 33 Another important aspect hampering the advances in scar research is the different phenotype, in which hypertrophic scars can occur. After surgery they appear rather localized and in single strands yielding them suitable for targeted injection therapy, while they appear rather diffuse and heterogenic after burns or secondary wound closure, yielding them more suitable for widespread topical treatment with the possibility of injection therapy in single localized strands. These different occurrences as well as the different stages of ripeness of scars, that also influence the choice of therapy have resulted in standardized human studies having been performed insufficiently.
Mustoe et al suggest "a move to a more evidence-based approach in scar management", already in 2002. 34 Ogawa in 2010 35  This review shall serve as an update to again emphasize the critical need for evidence for this considerable condition.

| METHODS
A systematic review of the literature has been conducted. The methods resembled those used in this previous study, 36  AND "scar" AND ("treatment" OR "therapy" OR "scar revision" OR "pressure garment")]. To minimize the risk of missing relevant data, the MeSH-term ["Cicatrix, Hypertrophic/therapy"[Mesh]] has been used, additionally. Finally, in a third step, the appropriate articles have been analyzed for their scientific value, since only level I evidence articles were to be included. For this assessment, the evidence rating scales of the American Society of Plastic Surgeons (ASPS) have been used, 37 with the additional refinement of only deeming a study the level of evidence I ("high-quality, multi-centered or single-centered, randomized controlled trial with adequate power"), if the study protocol (a) included a cohort of at least 15 scars per treatment arm, (b) there was a followup of at least 12 weeks after start of the therapy, (c) the dropout rate was below 20% after 12 weeks, and (d) the control group consisted of none, a placebo, or an intervention that is recommended "without restriction" in the most current guideline for the treatment of hypertrophic scars by the working group of scientific medical societies e.V.

| Reference selection and inclusion criteria
(AWMF; 4). 39 The remaining studies were included and have been analyzed for their specific content.

| Qualitative search results
A total of nine studies were found suitable for this analysis. Of these nine studies, four compared the effect of intralesional injections, whereof three analyzed triamcinolone (TAC) injection vs intralesional TAC + 5-FU [40][41][42] and one investigated the intralesional injection of TAC opposed to that of verapamil. 43 Two studies investigated the effect of topical ointments (silicone 44 and enalapril 45 ), one that of short-term massage, 46 one that of topical silicone dressing, that of pressure therapy and a combination thereof. 47 The ninth study protocol investigated the effect of CO 2 -laser on hypertrophic scars. 48 While the study by Nedelec et al investigating the effect of shortterm massage (5 minutes for three times a week) was the only one yielding no significantly different results, 46 the other studies could achieve significant improvements in one group over the other(s) in some investigated scar properties.
Study summaries are given in Table 1, while Table 2 summarizes an overview of the significant results.  Another relevant aspect in these studies is the fact, that they acquired large cohorts (150, 41  Also longer follow-up and intraindividual protocols are recommended.

| Intralesional Injection
Another possible target substance, which has shown similar regulating mechanisms, is botulinum toxin type A. 51 Its theoretical mechanism of action has been shown to be in an inhibition of TGF-b1 as well as in an increase of the JNK phosphorylation leading to reduced fibroblast proliferation and production of profibrotic factors, among others. 52 No level of evidence I studies have investigated this substance to date.

| Topical therapy
Three of the analyzed studies used intraindividual approaches to determine the effect of (a) short-term massage (+ lotion) vs.
lotion alone in 70 patients, 46 (b) enalapril (ACE inhibitor) ointment vs. placebo in 30 patients, 45 and (c) silicone gel vs. placebo in 38 patients. 44 While the short-term massage (5 minutes, 3 times a week for 12 weeks) yielded no significant difference in the groups, the Enalapril ointment (twice daily for 6 months) led to smaller scars with lower itching scores, yet no results for the indicatedly measured thickness was given. A major downside of these two studies is the fact, that no prolonged follow-up after the last intervention was indicated, and in the case of Mohammadi et al, no dropout was mentioned, either. 45 The third intraindividual study compared topical silicone gel vs placebo for 16 weeks. 44 In this study, a significant amelioration of pigmentation, vascularity, pliability, and itchiness could be achieved; however, no pain difference was reported. One downside is the fact, that there was, again, no prolonged follow-up after the treatment. The fourth study investigating topical treatments used four groups to determine the effect of silicone gel dressing and pressure therapy alone, a combination thereof, and no treatment. 47 The treatment lasted for 6 months and an additional month was used as follow-up period. A combination of the two means was deemed superior to pressure therapy alone, silicone gel dressing, and no treatment, in this order. Yet, silicone gel dressing was superior when it comes to the validation of pain and pruritus as compared to pressure therapy. Color, thickness, VSS and a Visual Analogue Scale for pain were assessed in this study.
Concluding, topical therapeutics represent a valuable, noninvasive approach to the treatment of hypertrophic scars. While silicone gel (dressings) seem to have their effect primarily on pruritus (and pain), the exact effect of physical approaches and other ointments is still to be validated in high-quality randomized-controlled trials, focusing on the same end points.

| CO 2 -Laser
One study used the ablative CO 2 -laser to treat hypertrophic scars. 48 Forty-eight scars in 36 patients were treated with a total of three sessions of CO 2 -laser every 4 to 6 weeks with an energy of 30 W. Thirty-two scars in the same patients were used as negative controls. Generally spoken, in this study, a significant decrease in scar thickness, pain sensation, erythema, and pigmentation was observed with a nonsignificant, yet present increase in skin elasticity. The control group also yielded an amelioration in the described properties; yet the amelioration curve was steeper in the laser group. With "worse" start points, however, a randomization bias cannot be excluded in this study. In our opinion, a clear recommendation for the use of CO 2 -laser for the treatment of hypertrophic scars cannot be drawn from this study. The use of amniotic membrane as skin graft dressing for one has been shown to be correlated with rapid reepithelialization and wound healing. 56,57 With delayed wound healing being a relevant factor in the occurrence of hypertrophic scars, 58 amniotic membrane has been shown to reduce the occurrence of hypertrophic scars post-burn. 59 A thorough understanding of pathomechanism and mechanisms of action, and well designed, prospective studies could encourage a standardized use of preventative measures after (burn) surgery to prevent/reduce the occurrence of hypertrophic scars.

| Limitations
Finally, TGF-β1 remains one of the key targets in the treatment of hypertrophic scars. 60 Accordingly, a specific block of TGF-β1 by, for example, antioxidants or Shikonin, an active component extracted from Chinese herbs, represent a promising approach against hypertrophic scars. These specific blockings have been proven effective in vitro already. 61,62 Not only the substances per se, also the way of application is a determining factor. The stratum corneum of the epidermis usually constitutes an effective barrier for many substances of large molecule size, being an obstacle for active ingredients to get to their place of action. The use of liposomes could therefore represent a valid strategy to properly deliver active substances, rendering painful injections redundant. 63 The application of papain, a cysteine protease from the papaya fruit, has shown promising results as active agent against hypertrophic scars in vitro and in an animal model when applied within liposomes. 64 These represent only a few aspects to consider when looking for the best strategy in scar treatment, encouraging clinicians as well as basic researchers to pool their competences and reduce the burden of hypertrophic scars.

| CONCLUSION
According to this review, it can be summarized, that even in 2019 there are still few studies, fulfilling the criteria of being rated as level of evidence 1. The last 10 years produced only nine studies in that category.

ACKNOWLEDGMENTS
No funding occurred in the preparation of this article.