Keloid scars are benign skin tumors that are unique to human beings. They occur secondary to skin injury, with localized protruding keloid scars that are itchy, painful and uncomfortable, and are constantly expanding into the surrounding normal skin without self-limiting or degenerative changes. Currently, although there are various treatments for keloid scars, the results are not satisfactory. After more than a hundred years of development, there are various surgical procedures, such as excision and suturing, skin grafting and flap grafting, etc. However, the recurrence rate of pathological scarring, especially keloid, is high when surgery alone is applied. Non-surgical treatment is relatively less invasive, simpler and easier to treat. The main treatment methods are the following: physical therapy (compression therapy, radiotherapy, laser therapy), traditional Chinese medicine (Centella asiatica glycosides, bitter ginseng alkaloids), and pharmacotherapy (corticosteroids, antimetabolites, retinoids, silicone, etc.). In conclusion, each of these methods has its own strengths and weaknesses. Pingyangmycin is an antitumor drug, which can inhibit the proliferation of vascular endothelial cells and clinically treat various hemangiomas, inhibit the blood supply of scar tissue and cause necrosis and liquefaction of scar tissue, in addition, Pingyangmycin can also inhibit the activity of collagen synthase, so that collagen deposition can be controlled. In order to further search for a convenient, fast, safe and effective method to treat keloid, we used local injection of Pingyangmycin to treat keloid, and observed the histological changes of keloid treated by local injection of Pingyangmycin using histological and histochemical methods to provide theoretical basis for clinical treatment of keloid. Clinical treatment Group A: Routine blood tests and chest fluoroscopy were performed to determine whether the patients could tolerate the treatment. Preparation of Pingyangmycin solution: Pingyangmycin 8mg,dissolved in 2% lidocaine hydrochloride injection 5ml diluted to present a colorless and transparent liquid in appearance, prepared before treatment and the remaining part discarded after treatment. Treatment method: Disinfect the area with 1% iodine, use a 5ml syringe, hold a No.5 needle, enter the needle along the base of the lesion, the tip of the needle reaches the middle of the scar, and inject evenly from the site of the needle in a fan to change the skin of the scar lesion from red to pale white. After the injection, pressure is applied to the injection point so that no drug leaks out of the needle hole. If the scar is large, multiple sites can be injected at one time, and the scar is 1.5cm above the skin surface, the scar will be surgically excised first, and the maximum dose of Pingyangmycin injection will be no more than 0.2mg/kg body mass for one injection, once in 2 weeks, 2 times for one course, and the results will be judged after 3 courses of treatment. Group B: Prednisolone acetate injection 125mg dissolved in 2% lidocaine hydrochloride injection 5ml, the base of the skin lesion closed treatment. The injection method and course of treatment were the same as that of the treatment group. Histological sections All keloid specimens were obtained from patients who voluntarily underwent keloid injection treatment in our hospital. 74 cases in group A, 88 keloid scars; 52 cases in group B, 56 keloid scars. All patients were free from infections and ulcers, and were free from infectious, immune, genetic and skin diseases. The purpose of sampling was explained to the patients and consent was obtained before sampling. After successful anesthesia, 1% vital iodine was used to disinfect the trunk and extremities, and the keloid was excised from the lesion site, of which a 10 mm × 10 mm size scar was placed in 10% formaldehyde. Section production Routine HE staining: extraction, fixation, dehydration, transparency, embedding, sectioning, HE staining and sealing histochemical staining (picric acid 2 Sirius scarlet staining): ① Dewaxing, wax sections were routinely dewaxed to water; ② Staining, azurite blue solution for 10 min; distilled water washing 3 times; Sirius scarlet saturated picric acid solution beam for 15- -30min; anhydrous ethanol direct differentiation and dehydration; ③ dehydration, transparency, sealing, xylene transparent, neutral resin sealing. The results were judged to appear green stained for type III collagen fibers, red or yellow for type I collagen fibers. The HPIAS22000 high-definition color pathology graphic report management system (Tongji Thousand Screen Imaging Company) was used for quantitative analysis of picric acid-aspirate scarlet staining. Five complete and non-overlapping high-powered fields of view (×400) were randomly selected for each section, and the positive reaction area and the total area of all fibers under each field of view were measured to calculate the positive area rate. The average of the positive area ratio of 5 fields of view for each case was used as the measurement value for that case (positive area ratio = total area of positive reaction per unit area/total area of fibers per unit area × 100%). The number of fibroblasts in the scar tissue of group B was significantly increased, and the cells were distributed near the collagen fibers, and the collagen fibers were densely packed and disorganized, with glassy changes in some tissues. The histochemical staining of group A showed that the scar tissue was dominated by scattered green slender type III collagen fibers, with fewer red and yellow thick type I collagen fibers. In contrast, in group B, the hyperplastic scar tissues showed dense distribution of red or yellow thick type I collagen fibers in bundles with disorganized arrangement and uneven density distribution; type III collagen fibers were less in number and were scattered around type I collagen in a sparse network. By one-way ANOVA, there was a significant difference between groups (P<0.01. By q test, there was a significant difference in the positive area rate of type III collagen fibers between group A and group B.