2.1 Study Population
All granulomatous lobular mastitis (GLM) patients hospitalized in the Department of Mammary Surgery at the Renmin Hospital of Wuhan University from January 1, 2017, to December 31, 2017, were reviewed. The patients complained of “breast lump or pain”. The clinical manifestations, previous medical history, and laboratory examinations were consistent with the characteristics of GLM. The formation of hypoechoic masses or abscesses of breasts (according to GLM characteristics) was found under ultrasound. The ultrasound-guided breast needle biopsy was performed and sent for pathological examination. After the pathological diagnosis was GLM, minimally invasive surgery was performed. The specimens removed from the surgery were sent to routine pathological examination again. The pathological results were reviewed by two pathologists. After the diagnosis of GLM, dressing change, lesion washing, and oral steroid treatment were performed. Among them, minimally invasive surgery and subsequent dressing changes, lesion flushing and oral steroid therapy together constitute the "minimally invasive comprehensive treatment" described herein. The whole workflow is shown in Fig 1.
Inclusion criteria: female; patients who were clinically diagnosed and pathologically confirmed with GLM; and patients undergoing minimally invasive comprehensive treatment.
Exclusion criteria: patients with tuberculous mastitis; patients with fat necrosis; patients with sarcoidosis; patients with periductal mastitis; patients with inflammation due to lactation or pregnancy; patients with other possible causes of mammary inflammation or granulomatous changes.
2.2 Minimally invasive comprehensive treatment
All GLM patients received minimally invasive comprehensive treatment, which was operated with an ultrasound-guided Mammotome minimally invasive biopsy system. We placed the indwelling hoses in the lesion during the procedure and then flushed the lesion via those indwelling hoses. After the operation, we regularly changed the dressing and irrigated the lesion. The oral steroid was administered after confirming that the histopathology of the patient was found to be GLM. The specific treatment methods were as follows:
2.2.1 Minimally invasive operation
After determining the location, size, and extent of the lesion by ultrasound examination, the appropriate puncture points were selected. Local anesthesia was administered with 1% lidocaine and adrenaline to the tissues surrounding the puncture point, puncture path, and lesion. A small incision of about 30 mm long was made with a sharp blade at the puncture point. The Mammotome rotary scalpel was inserted through the skin incision into the bottom of the lesion under the guidance of ultrasound.
The rotary scalpel was then adjusted so that the groove was located behind the lesion and circumcision and aspiration were performed on the lesion. For patients with small lesions, complete excision was attempted. In patients with large lesions, many small pus cavities were often formed or merged into larger pus cavities. At this time, the focus of the minimally invasive operation was to try to open each pus cavity and expel internal pus, rather than pursuing a complete excision of the lesion to avoid causing greater damage to the patient’s breast structure. For lesions with complex internal structures, it was difficult to completely remove the pus, and we tried to insert a rotary scalpel from multiple puncture points for better surgical results.
2.2.2 Intraoperative placement of hoses and irrigation of lesions
After excision of the lesion or opening of the pus cavity, the indwelling hoses were inserted into the lesion from the surface of the breast skin. The lengths and specifications of the indwelling hoses were selected according to the depth and scope of the lesion. Two to six indwelling hoses were used for unilateral breast lesions. The number of indwelling hoses could be increased appropriately for lesions with large scope and complicated internal structure, so as to achieve the effect of adequate drainage and flushing of the lesion. The hoses could also be inserted into the lesion from different directions to achieve a better irrigation effect. Normal saline was injected into the indwelling hoses to confirm that the hoses were unobstructed, and then the lesions were irrigated with hydrogen peroxide, metronidazole, and dexamethasone (Fig 2).
2.2.3 Postoperative dressing change and lesion lavage
After the operation, each patient underwent regular dressing changes and lesion flushing. Here, the 2 to 6 indwelling hoses placed during surgery were used. One end of these hoses led to the lesion and the other was left on the skin surface. Because the lesion was the pus cavity or the surgical cavity remaining after the substantial lesion was surgically removed, so the lesion ends of the indwelling hoses were connected. When the drug was injected into the lesion from one indwelling hose, the drug can flow out of the other hoses, and the drug was continuously injected to play the effect of flushing the lesion. During rinsing, some pus, necrotic tissue, and newly generated pus that were not easily discharged in the cavity of the lesion can be flushed out of the lesion together, which was beneficial to the recovery and healing of the lesion.
Prior to every irrigation, normal saline was injected into the indwelling hoses to confirm the hose patency, and then the lesion was washed with iodophor, metronidazole, and dexamethasone. During the first week after surgery, dressing changes and irrigations were performed daily and then the frequency was gradually reduced according to the patient’s recovery. When the inflammation of the patient’s lesion subsided and no new pus was being produced, the irrigation fluid became clear, and the mass gradually softened and disappeared, at which point the indwelling hose can be gradually removed. All indwelling hoses were removed from 2 weeks to 1 month after surgery.
2.2.4 Oral steroid treatment
After each patient was histopathologically confirmed as GLM, oral prednisone was prescribed at a dose of 15 mg/day in the first week after surgery, 10 mg/day in the second week, and 5 mg/day in the third week and thereafter. The medication was discontinued when the patient was fully recovered or unacceptable side effects occurred.
2.3 Data Collection
Patients’ information, including age, clinical features, image characteristics, histopathological findings, mastitis history, treatment process, operative technique, recurrence, and esthetic effect, was analyzed. Disease recurrence was defined as the reappearance of a mass, abscess, or fistula on the ipsilateral breast. All patients were guided by an independent research nurse to evaluate cosmetic outcomes before and after minimally invasive comprehensive treatment using a semi-quantitative visual simulation scale (3 points, excellent; 2 points, good; 1 point, acceptable; and 0 points, poor), and all patients were divided into 2 groups according to whether they had a history of breast surgery in the hospital before this minimally invasive comprehensive treatment.
2.4 Statistical Analysis
Statistical analysis was performed with SPSS 24.0 software. The results were reported as mean ± standard deviation for the continuous variables. A Wilcoxon rank-sum test for two independent samples was used to compare the differences in cosmetic scores between patients with or without previous breast surgical history.