Abstract
Background
Idiopathic granulomatous mastitis (IGM) is a chronic inflammatory breast disease with an unclear etiology. Patients with IGM remain at risk of recurrence even after clinical cure. We assessed the long-term clinical outcomes of segmental mastectomy paired with vacuum sealing drainage (VSD)-based integrated wound management for IGM patients eligible for surgical intervention.
Methods
Patients were divided into two groups according to the clinical treatment protocol: 87 patients underwent segmental mastectomy combined with VSD (VSD comprehensive group), and 63 underwent segmental mastectomy combined with conventional drainage (conventional drainage group). Postoperative pain score, number of dressing changes, incision healing time, length of hospital stay, hospitalization cost, breast morphology, and 2-year recurrence rate were compared between groups.
Results
Compared with the conventional drainage group, the VSD comprehensive group had a significantly lower 2-year recurrence rate [4.59% vs. 20.63%; odds ratio (OR) =0.185, 95% confidence interval (CI): 0.057–0.600; P=0.003] and a significantly higher rate of excellent breast morphology (68.97% vs. 49.21%; OR =2.294; 95% CI: 1.173–4.487; P=0.02). Significant intergroup differences were observed in pain score (5.989±0.896 vs. 3.984±0.813, P<0.001), number of dressing changes (14.862±3.638 vs. 10.730±5.074, P<0.001), incision healing time (38.081±7.007 vs. 19.048±12.681 days, P<0.001), length of hospital stay (9.931±1.797 vs. 8.984±2.703 days, P=0.01), hospitalization cost (¥12,186.471±1,348.616 vs. ¥11,628.571±1,274.002, P=0.01).
Conclusions
Segmental mastectomy paired with VSD-based integrated wound management was associated with improved breast morphology and a lower 2-year postoperative recurrence rate in patients with IGM, although early postoperative recovery was slower and hospitalization cost was higher.
Keywords: Idiopathic granulomatous mastitis (IGM), vacuum sealing drainage (VSD), segmental mastectomy, 2-year recurrence rate
Highlight box.
Key findings
• Segmental mastectomy paired with vacuum sealing drainage (VSD)-based integrated wound management was associated with a lower 2-year local recurrence rate [4.59% vs. 20.63%; odds ratio (OR) =0.185, 95% confidence interval (CI): 0.057–0.600; P=0.003] in patients with idiopathic granulomatous mastitis (IGM). Comprehensive wound management using VSD was also associated with better breast cosmetic outcomes (68.97% vs. 49.21%; OR =2.294; 95% CI: 1.173–4.487; P=0.02), although postoperative recovery was slower and hospitalization costs were higher.
What is known and what is new?
• IGM mainly affects women of reproductive age. Surgery alone is associated with a considerable recurrence rate and may impair breast appearance.
• This study compared the long-term therapeutic efficacy of VSD-assisted comprehensive wound management versus conventional drainage following segmental mastectomy in patients with IGM.
What is the implication, and what should change now?
• Segmental mastectomy combined with comprehensive VSD wound management may be considered in selected patients after shared decision-making, particularly when long-term recurrence reduction and breast cosmesis are prioritized.
Introduction
Idiopathic granulomatous mastitis (IGM) is a relatively rare chronic inflammatory breast disease. In recent years, its incidence has shown a marked upward trend (1). IGM typically affects reproductive-aged women and is strongly associated with a history of pregnancy, parturition, and lactation, whereas male involvement is exceedingly rare (2,3). Pathologically, IGM is characterized mainly by multifocal noncaseating granulomas composed primarily of epithelioid cells, Langhans giant cells, neutrophils, and lymphocytes, and it may be accompanied by microabscesses (4,5). The main clinical manifestations of IGM include breast pain and swelling, erythematous and swollen breast skin, nipple retraction, nipple discharge, breast fistula, sinus tract formation, and enlarged axillary lymph nodes (6). Current management of IGM is individualized, with therapeutic decisions informed by both disease stage and clinical presentation. Therapeutic strategies encompass expectant management, pharmacological agents (including corticosteroids, methotrexate, other immunosuppressants, and antibiotics when indicated), minimally invasive interventions such as image-guided drainage, and definitive surgical resection (5,7). However, segmental breast resection can be extensive, which may cause breast deformity and affect patients’ appearance and mental health (8). Drug therapy has certain advantages, including fewer adverse effects and less trauma; however, it requires a long treatment period, patient compliance is often poor, and the recurrence rate remains high (9). Therefore, identifying a safe, effective, and minimally invasive treatment strategy remains an important focus in the clinical management of IGM.
Vacuum sealing drainage (VSD), a wound treatment technique, has been widely used for various acute and chronic wounds since its introduction in the 1990s and has achieved favorable therapeutic effects (10). The comprehensive wound management strategy—integrating VSD with irrigation, open wound care, repeated dressing changes, and delayed secondary closure—operates through a synergistic triad of debridement, promotion, and repair. Specifically, negative pressure combined with irrigation eliminates exudate and bacteria; negative pressure stimulates granulation tissue proliferation; and delayed closure ensures wound bed maturation. This sequential approach effectively prevents infection and facilitates definitive wound healing (11,12). Segmental mastectomy effectively removes lesions through precise excision of diseased tissue, thereby reducing the risk of recurrence. The combination of these two methods is expected to completely remove lesions while reducing surgical trauma, promoting wound healing, maximizing preservation of breast shape and function, and improving patients’ quality of life. Therefore, segmental mastectomy combined with VSD comprehensive wound management strategy has considerable clinical significance and application value in reducing recurrence, maintaining breast shape, and improving prognosis. We present this article in accordance with the STROBE reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0372/rc).
Methods
Study population
A retrospective cohort study was conducted to investigate the clinical efficacy of segmental mastectomy combined with VSD-assisted comprehensive wound management in patients with IGM. Patients diagnosed with IGM who were treated in the Department of Breast Surgery of Gansu Provincial Hospital between January 2018 and July 2023 were enrolled according to predefined inclusion and exclusion criteria. Patients were categorized according to postoperative drainage and wound management modalities. Because drainage strategy was not randomly assigned, the potential for selection bias was considered when interpreting the results. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Gansu Provincial Hospital (No. 2025-528). The requirement for individual informed consent for this retrospective analysis was waived. All patient data were kept confidential.
Inclusion criteria were as follows: (I) receipt of standardized, stage-adapted treatment in accordance with the 2021 International Multidisciplinary Consensus (5) and the 2024 Chinese Expert Consensus on granulomatous lobular mastitis (GLM) (13), with fulfillment of the predefined surgical indications; (II) age 18–60 years; (III) segmental mastectomy combined with VSD-assisted comprehensive wound management or conventional drainage performed under general anesthesia; and (IV) a follow-up duration of at least 2 years.
Exclusion criteria were as follows: (I) concurrent breast malignancy or other systemic malignancies; (II) severe systemic disorders that could impair wound healing or postoperative recovery; (III) pregnancy or lactation with surgical contraindications; (IV) prior nonstandard hormone therapy, immunosuppressive therapy, or invasive procedures performed at other institutions before admission; and (V) incomplete clinical or imaging data, follow-up of <2 years, or loss to follow-up.
Clinical management pathway
All patients were managed according to a prespecified clinical algorithm (Figure 1), which was developed in accordance with the 2021 International Multidisciplinary Consensus on the Management of Granulomatous Lobular Mastitis (5) and the 2024 Chinese Expert Consensus on the Pathological Diagnosis of Granulomatous Lobular Mastitis (13). The pathway consisted of six sequential steps:
Figure 1.
Standardized clinical management pathway for IGM in this cohort. The pathway was formulated based on the 2021 International Multidisciplinary Consensus on the Management of Granulomatous Lobular Mastitis and the 2024 Chinese Expert Consensus on the Pathological Diagnosis of Granulomatous Lobular Mastitis. CR, complete response; IGM, idiopathic granulomatous mastitis; PD, progressive disease; PR, partial response; US, ultrasound.
Diagnostic criteria and etiologic exclusion for IGM
According to the 2021 International Multidisciplinary Consensus, a diagnosis of IGM required typical histopathologic features and complete exclusion of secondary causes. GLM confirmed by core-needle biopsy was histologically defined by lobule-centered, noncaseating epithelioid granulomas with mixed lymphocytic, plasma cell, and neutrophilic infiltration; cases showing carcinoma, atypical hyperplasia, or caseous necrosis were excluded (Figure 2).
Figure 2.
Pathological manifestations of IGM (HE staining, ×20). IGM, idiopathic granulomatous mastitis; HE, hematoxylin and eosin.
All enrolled patients underwent standardized etiologic screening to exclude secondary causes of granulomatous mastitis. Acid-fast staining, Mycobacterium tuberculosis culture, and tuberculosis polymerase chain reaction (PCR) assays yielded negative results in all patients. Fungal smears and cultures, routine bacterial cultures, and Corynebacterium sequencing were also negative. Cystic neutrophilic granulomatous mastitis (CNGM) was excluded, and histopathologic examination showed no evidence of foreign-body granulomas or granulomatous lesions secondary to milk stasis. In addition, no patient had secondary hyperprolactinemia attributable to long-term oral contraceptive use, psychotropic medication use, or pituitary adenoma. Patients with the characteristic histopathologic features of GLM and negative results on all of the aforementioned investigations were diagnosed with IGM and enrolled in the study.
Pretreatment clinical severity staging
A four-stage classification system for IGM proposed by the 2021 International Multidisciplinary Consensus (5) was used.
Stage I (self-limited): mass <5 cm, with mild tenderness and no erythema, abscess, or lymphadenopathy.
Stage II (congestive): mass ≥5 cm with pain, hyperemia, and swelling; lesion confined to a single quadrant.
Stage III (abscess): involvement of 2 quadrants with confirmed abscesses; axillary lymphadenopathy may be present, whereas chronic sinuses are absent.
Stage IV (complex): involvement of ≥3 quadrants with multiple abscesses, ulcers, chronic fistulas, and persistent discharge; arthralgia or erythema nodosum may be present.
Initial systemic treatment protocol (consensus-recommended first-line therapy and standardized local protocol for this cohort)
Before corticosteroid initiation, routine bacterial culture of pus was performed in all patients. Patients with confirmed infection first received targeted antibiotics, and corticosteroid therapy was initiated only after complete infection control. Stage-stratified treatment was administered as follows:
Stage I (self-limited stage): Routine corticosteroid therapy was not indicated. Regular clinical and ultrasonographic follow-up was performed.
Stages II/III (congestive/abscess stages): Oral methylprednisolone was administered at 20 mg once each morning according to a standardized 8-week tapering regimen: 20 mg/day for the first 4 weeks, 10 mg/day during weeks 5–6, and 5 mg/day during weeks 7–8. Treatment was discontinued at the end of Week 8, when therapeutic response was assessed. For patients with concomitant abscesses, ultrasound-guided puncture and complete aspiration of pus were performed before corticosteroid initiation. The abscess cavity was then irrigated with normal saline once weekly until purulent drainage ceased, while oral methylprednisolone was administered concurrently. Long-term corticosteroid monotherapy was not recommended for lesions with abscess formation.
Stage IV: Oral methylprednisolone and repeated ultrasound-guided aspiration were used initially to control inflammation and abscesses, followed by surgical resection. Methylprednisolone was used only as a preoperative bridge because aspiration alone provided temporary relief rather than definitive treatment.
Standardized therapeutic response evaluation
Treatment response was reassessed every 4 weeks by palpation and breast ultrasonography, with a final assessment at Week 8 according to standard IGM trial criteria:
Complete response (CR): complete disappearance of all lesions (masses, abscesses, and fistulas), with resolution of pain and erythema.
Partial response (PR): ≥50% reduction in lesion volume without new lesions, accompanied by substantial symptomatic improvement.
Poor response/progressive disease: <50% reduction in lesion volume, any increase in lesion size, or development of new abscesses/fistulas.
Definitions of refractory and recurrent disease
Refractory disease was defined as failure to achieve a PR or CR after 8 weeks of standard-dose methylprednisolone (20 mg/day), with imaging showing <50% reduction in maximal lesion diameter or documented disease progression.
Recurrence was defined as the development of pathologically confirmed active lesions at the primary site or elsewhere in the ipsilateral or contralateral breast within 6 months after clinical cure, defined as complete clinical and sonographic resolution of all target lesions.
Surgical indications
Surgical intervention was offered to patients who met any of the following criteria: (I) refractory disease after standardized corticosteroid therapy (lesion reduction <50% or disease progression despite adequate treatment); (II) recurrent disease; (III) stage IV disease with sinuses and fistulas or involvement of ≥3 breast quadrants; or (IV) lack of response to corticosteroid therapy or intolerance to corticosteroid-related adverse effects.
Group assignment and treatment allocation
Group assignment was based on preoperative disease severity features. The conventional drainage group included patients with lesions ≤5 cm confined to ≤2 quadrants and without sinus tracts, fistulas, ulceration, or multiple abscesses. The VSD comprehensive group included patients with any of the following: lesion size >5 cm, involvement of ≥3 quadrants, sinus/fistula formation, skin ulceration, or multiple abscesses.
Surgical procedures
Preoperative preparation
A detailed medical history, including previous breast disease, drug allergies, and menstrual and reproductive history, was obtained by the resident physician. Comprehensive examinations were performed, including routine blood testing, coagulation function testing, liver and kidney function testing, electrocardiography, chest radiography, breast ultrasonography, and abdominal color Doppler ultrasonography. Breast magnetic resonance imaging was also performed to determine the extent, location, and depth of the lesion. The surgeon or first assistant explained the surgical procedure, expected outcomes, potential risks, and corresponding management measures to the patient and family, and surgical informed consent was obtained.
Surgical procedure
After the patient entered the operating room, preoperative preparation was completed and general anesthesia was administered. After routine disinfection and draping, an appropriate surgical incision was selected according to the preoperatively determined lesion extent. A periareolar arc incision was made if the lesion was adjacent to the areola, whereas a radial incision was used if the lesion was located in the peripheral breast. After the skin and subcutaneous tissue were incised, the breast tissue was bluntly dissected, and the lesion segment was resected with 1–2 cm of surrounding normal breast tissue using a ConMed System 2450 electrosurgical unit (ConMed Corporation, USA; Model: 60-2450-220). If the wound was extensive, a surrounding breast gland flap was transposed to maintain the breast contour. After lesion resection, meticulous hemostasis was achieved using suture ligation and electrocoagulation. The wound surface was repeatedly irrigated with normal saline and hydrogen peroxide to remove residual tissue debris and blood. In the VSD group, part of the incision was closed with interrupted sutures, and a VSD negative-pressure suction device was placed through the remaining 3–4 cm incision. In the conventional drainage group, a conventional negative-pressure drainage tube was inserted before complete interrupted closure of the incision.
Placement of the VSD negative-pressure drainage device
After thorough hemostasis and irrigation, the VSD (Chongqing Qingshi Biotechnology Co., Ltd., Chongqing, China, Type I-PVA) material was trimmed by the assistant to match the size and contour of the wound. The trimmed VSD material was placed into the wound to ensure complete coverage and to avoid dead space. A biological semipermeable membrane was then used to seal the wound and VSD material completely, thereby creating a closed environment. The negative-pressure suction device was connected and checked by the nurse, and drainage through the catheter was observed. After the patient returned safely to the ward, the bedside negative-pressure suction device was connected and adjusted to 0.02–0.03 MPa. Postoperatively, the VSD was irrigated with 1,500 mL of normal saline daily, and the wound condition, bleeding or exudation, patency of the VSD system, and color, volume, and characteristics of drainage fluid were monitored. If obvious bleeding was observed during wound irrigation, the dressing was removed promptly, the wound was inspected, and hemostasis was performed. If the drainage fluid was bright red and gradually increased in volume, bleeding was suspected and close observation with appropriate intervention was required. If the drainage fluid was turbid or foul-smelling, infection was suspected, and bacterial culture and drug sensitivity testing were performed to guide antimicrobial treatment. The VSD material and drainage device were replaced according to the wound condition, usually every 4–5 days, under strict aseptic technique. Wound healing status, including granulation tissue growth and wound contraction, was monitored. Postoperatively, patients were advised to increase intake of high-protein, high-calorie, and high-vitamin foods to promote wound healing. When the wound diameter decreased to less than 2 cm, suturing was performed. The detailed process is shown in Figure 3.
Figure 3.
The application of VSD in a patient with IGM. (A) Preoperative ulceration of the skin overlying the lesion. (B) The surgical cavity underwent complete drainage and irrigation via VSD. (C) One week after VSD and irrigation, healthy granulation tissue was observed within the surgical cavity. (D) At 32 days postoperatively, the patient’s surgical cavity had gradually achieved complete healing. IGM, idiopathic granulomatous mastitis; VSD, vacuum sealing drainage.
Conventional drainage
After hemostasis and thorough wound irrigation, a 16-Fr negative-pressure drainage tube [HuBei Kangquan Medical Technology Co., Ltd., Yichang, China, Model: SY-16Fr(H)] was inserted, and the insertion depth was determined according to wound size. Postoperatively, the characteristics of the drainage fluid, including color, volume, and properties, were observed and recorded. The tube was removed when the drainage volume was less than 10 mL for three consecutive days. Bleeding or infection was managed similarly to the VSD group. If incision nonhealing or exudation occurred, part of the incision sutures was removed, the wound was packed with gauze, and dressing changes were performed until healing was achieved.
Data collection
Data from enrolled patients with GLM were retrospectively collected, including onset-related information, previous treatment records, and postoperative recovery status. The clinical data consisted of four parts: (I) General patient information, including age, disease course, breastfeeding history, pregnancy history, nipple inversion, nipple discharge, contraceptive use, smoking history, breast trauma history, hyperprolactinemia, and other medical history. (II) Disease severity features: abscess formation, sinus/fistula formation, skin ulceration, involved breast quadrants, lesion dimension; disease severity was quantified using a composite score derived from four parameters: abscess count (0= none, 1= single, 2= multiple), sinus/fistula (0= absent, 2= present), skin ulceration (0= absent, 2= present), and involved quadrants (1= single, 2= multiple). Total scores ranged from 0 to 8. (III) Inflammatory markers: C-reactive protein (CRP), erythrocyte sedimentation rate (ESR). (IV) The primary outcome endpoint was 2-year recurrence rate and breast morphology. The secondary endpoints were: (i) pain score at the first postoperative dressing change; (ii) number of dressing changes; (iii) incision healing time; (iv) length of hospital stay, and (v) total hospitalization costs. Incision healing time was defined as the interval from surgery to complete wound healing as documented in the medical record. Breast morphology was graded as excellent, good, fair, or poor according to the postoperative cosmetic assessment recorded during follow-up. Two-year recurrence was defined as clinically documented local recurrence during the 24-month follow-up period.
Statistical analysis
Data from patients with GLM were statistically analyzed using SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as the mean ± standard deviation (SD). Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the F test. Between-group comparisons of continuous variables were performed using Student’s t-test when the assumptions of normality and homogeneity of variance were met. Categorical variables were summarized as frequencies and percentages, and between-group comparisons were performed using the chi-square test or Fisher’s exact test, as appropriate. Propensity score matching (PSM) was attempted to control for baseline confounding but yielded too few matched pairs for adequate power. Multivariable regression was also not feasible due to the limited number of recurrent events. Thus, we used standardized mean differences (SMDs) to assess baseline balance, with SMD <0.30 indicating acceptable comparability between groups. To assess the robustness of the findings, three sequential logistic regression models were constructed with hierarchical entry of covariates: Model 1 was unadjusted; Model 2 was adjusted for lesion dimension; and Model 3 was adjusted for both lesion dimension and disease severity. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs).
Results
Baseline characteristics
A total of 182 patients with GLM were initially enrolled. After applying the predefined inclusion and exclusion criteria, 32 patients were excluded, including 25 with identifiable secondary etiological factors, 4 lost to follow-up, and 3 with incomplete clinical data. Ultimately, 150 patients were included in the final analysis, with 87 in the VSD comprehensive group and 63 in the conventional drainage group. In addition, 10 patients with concurrent isolated hyperprolactinemia received standardized bromocriptine therapy, and serum prolactin levels normalized in all cases. Baseline disease severity was substantially greater in the VSD comprehensive group than in the conventional drainage group, as reflected by larger lesions (4.93±0.88 vs. 3.67±0.76 cm; SMD =1.52) and higher disease severity scores (4.97±1.65 vs. 1.51±0.64; SMD =2.61), higher proportions of multiple abscesses (39.08% vs. 0%), sinus/fistula formation (35.63% vs. 0%), skin ulceration (60.92% vs. 0%), and involvement of ≥3 quadrants (26.44% vs. 0%), as well as higher inflammatory marker levels (CRP: 15.75±5.70 vs. 12.81±4.44 mg/L, SMD =0.56; ESR: 27.76±7.08 vs. 19.75±3.68 mm/h, SMD =1.36). Demographic characteristics, including age, disease duration, breastfeeding history, and pregnancy history, were generally balanced (SMD <0.3). In contrast, SMDs for all severity-related variables exceeded 0.3, indicating clinically meaningful baseline imbalances between the two groups. Given these imbalances, sensitivity analyses were performed to assess the robustness of the treatment effect to potential confounding (Table 1).
Table 1. Preoperative baseline characteristics of patients.
| Characteristics | VSD comprehensive group | Conventional drainage group | P | SMD |
|---|---|---|---|---|
| Demographic and risk factors | ||||
| Age, years | 30.48±4.74 | 29.81±5.46 | 0.27 | 0.13 |
| Disease course, months | 10.62±4.40 | 10.16±5.27 | 0.08 | 0.10 |
| Breastfeeding history | 69 (79.31) | 42 (66.67) | 0.08 | 0.29 |
| Pregnancy history | 76 (87.35) | 48 (76.19) | 0.08 | 0.30 |
| Inversion of the nipple | 13 (14.94) | 6 (9.52) | 0.33 | 0.17 |
| Nipple discharge | 9 (10.34) | 12 (19.05) | 0.13 | 0.24 |
| Contraceptive history | 2 (2.30) | 2 (3.17) | >0.99† | 0.05 |
| History of breast trauma | 10 (11.49) | 4 (6.35) | 0.29 | 0.17 |
| Isolated hyperprolactinemia | 6 (6.90) | 4 (6.35) | >0.99† | 0.02 |
| Disease severity features | ||||
| Abscess formation | <0.001 | 1.13 | ||
| None | 27 (31.03) | 45 (71.43) | ||
| Single | 26 (29.89) | 18 (28.57) | ||
| Multiple | 34 (39.08) | 0 | ||
| Sinus/fistula formation | 31 (35.63) | 0 | <0.001 | 1.05 |
| Skin ulceration | 53 (60.92) | 0 | <0.001 | 1.77 |
| Involved breast quadrants | <0.001 | 0.95 | ||
| 1 quadrant | 29 (33.33) | 48 (76.19) | ||
| 2 quadrants | 35 (40.23) | 15 (23.81) | ||
| ≥3 quadrants | 23 (26.44) | 0 | ||
| Lesion dimension, cm | 4.93±0.88 | 3.67±0.76 | <0.001 | 1.52 |
| Disease severity score | 4.97±1.65 | 1.51±0.64 | <0.001 | 2.61 |
| Inflammatory markers | ||||
| CRP, mg/L | 15.75±5.70 | 12.81±4.44 | <0.001 | 0.56 |
| ESR, mm/h | 27.76±7.08 | 19.75±3.68 | <0.001 | 1.36 |
Data are presented as mean ± standard deviation or n (%). Between-group comparisons were performed using Student’s t-test for continuous variables and the chi-square test for categorical variables, unless otherwise indicated. †, Fisher’s exact test was used because more than 20% of the expected cell counts were less than 5. CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; SMD, standardized mean difference; VSD, vacuum sealing drainage.
Postoperative clinical data of patients with GLM
The VSD comprehensive group had a significantly higher postoperative pain score at the first dressing change, more dressing changes, longer incision healing time, longer hospital stay, and higher hospitalization cost than the conventional drainage group (all P<0.05; Table 2).
Table 2. Comparison of clinical data after surgery between the two groups.
| Characteristics | VSD comprehensive group | Conventional drainage group | t/Z | P |
|---|---|---|---|---|
| Pain score | 5.989±0.896 | 3.984±0.813 | 14.056 | <0.001 |
| Number of dressing changes | 14.862±3.638 | 10.730±5.074 | 5.810 | <0.001 |
| Incision healing time (days)† | 38.081±7.007 | 19.048±12.681 | 11.748 | <0.001 |
| Hospital stay (days) | 9.931±1.797 | 8.984±2.703 | 2.576 | 0.01 |
| Hospitalization costs (¥) | 12,186.471±1,348.616 | 11,628.571±1,274.002 | 2.559 | 0.01 |
Data are presented as mean ± standard deviation. Between-group comparisons were performed using Student’s t-test unless otherwise indicated. †, data for healing time were right-skewed and analyzed using Mann-Whitney U test. VSD, vacuum sealing drainage.
Postoperative breast morphology and 2-year recurrence rate in patients with GLM
The VSD comprehensive group had a higher rate of excellent breast morphology than the conventional drainage group (68.97% vs. 49.21%; OR =2.294; 95% CI: 1.173–4.487; P=0.02). The 2-year recurrence rate was significantly lower in the VSD comprehensive group than in the conventional drainage group (4.59% vs. 20.63%; OR =0.185; 95% CI: 0.057–0.600; P=0.003), corresponding to approximately 81.5% lower odds of recurrence with VSD treatment (Table 3). To evaluate the robustness of our findings against baseline imbalances, we performed hierarchical logistic regression with sequential adjustment for lesion diameter and disease severity (Table S1). The protective association of VSD-based management remained directionally consistent across all models (unadjusted OR =0.185; adjusted for lesion diameter: OR =0.277; fully adjusted: OR =0.286). The loss of statistical significance after full adjustment (P=0.27) was likely attributable to the limited statistical power consequent to the small number of recurrent events (n=17). Collectively, these sensitivity analyses support the robustness of the protective trend associated with VSD-based comprehensive management.
Table 3. Comparison of postoperative breast morphology and 2-year recurrence between the two groups.
| Outcome | VSD comprehensive group | Conventional drainage group | OR (95% CI) | P |
|---|---|---|---|---|
| Breast morphology | 60 (68.97) | 31 (49.21) | 2.294 (1.173–4.487) | 0.02 |
| 2-year recurrence | 4 (4.59) | 13 (20.63) | 0.185 (0.057–0.600) | 0.003 |
P values were calculated using Fisher’s exact test (two-sided). CI, confidence interval; OR, odds ratio; VSD, vacuum sealing drainage.
Discussion
In this retrospective cohort of 150 IGM patients (87 VSD vs. 63 conventional drainage), VSD-based integrated management was associated with higher breast morphology satisfaction (69.0% vs. 49.2%, P=0.02) and lower 2-year recurrence (4.6% vs. 20.6%, P=0.003), despite greater baseline severity (lesion size: 4.93 vs. 3.67 cm; severity score: 4.97 vs. 1.51). The sensitivity analyses demonstrated a directionally consistent protective association across all models, although statistical significance was attenuated after full adjustment. This attenuation is likely attributable to the limited number of recurrent events rather than to a true absence of effect, supporting the robustness of the protective trend while underscoring the need for validation in larger prospective cohorts.
GLM may represent an immune response triggered by retained milk and has been linked to autoimmunity, infection, hyperprolactinemia, smoking, and oral contraceptive use (14,15). Glucocorticoids are considered first-line therapy and may reduce the extent of surgery and improve breast cosmesis, but they can cause adverse effects, including hyperglycemia and osteoporosis (16,17). Although surgery, including local excision, segmental mastectomy, or mastectomy, allows prompt lesion removal, it may also result in breast deformity. More importantly, the long-term recurrence rate after surgery alone remains clinically relevant. Inadequate margins may increase recurrence risk, whereas extended resection may reduce recurrence at the expense of breast appearance. Accordingly, combined treatment strategies have been recommended (17-19). VSD applies controlled negative pressure to enhance tissue perfusion, promote granulation tissue formation, and reduce infection risk (18,19). Mechanistically, it may improve the wound microenvironment by reducing edema and dead space while supporting perfusion and tissue repair (20,21).
One notable finding was that the VSD comprehensive group had a lower recurrence rate despite substantially greater baseline disease severity. This apparently counterintuitive pattern may reflect the combined effects of the integrated VSD-based treatment model rather than negative-pressure drainage alone. The model included continuous negative-pressure drainage, irrigation, open wound management, repeated dressing changes, and delayed secondary closure, providing a systematic approach to complex lesions in severe IGM. First, negative pressure combined with irrigation may facilitate removal of deep necrotic tissue, inflammatory exudate, and bacterial biofilm while diluting local proinflammatory cytokines, thereby disrupting the cycle of inflammation, necrosis, and infection (20,21). Second, repeated dressing changes may function as gradual debridement, allowing stepwise removal of necrotic tissue while preserving newly formed granulation tissue. In addition, open wound management permits direct assessment of the wound bed and adjustment of negative-pressure and irrigation parameters, facilitating individualized treatment (22). Finally, delayed secondary closure follows the principle of “infection control first”, with closure performed only after the wound bed is free of residual infection. This approach may reduce dead space and promote the formation of well-vascularized, structurally robust healing tissue that is less susceptible to recurrence (23,24). Taken together, these components may help explain why patients with more severe disease in the VSD comprehensive group nevertheless had a lower recurrence rate. This interpretation is consistent with the sensitivity analyses, in which the association between VSD and lower recurrence remained directionally consistent after adjustment for lesion size and disease severity, suggesting that the observed association may reflect effective management of severe lesions rather than a direct anti-recurrence effect of negative pressure alone.
The rate of satisfactory breast morphology was higher in the VSD comprehensive group than in the conventional drainage group (68.97% vs. 49.21%; P=0.02). This difference may be related to differences in incision management and wound healing between the two approaches. In the conventional drainage group, the incision was primarily closed with a drainage tube left in situ. Although this approach allowed early wound closure, the tube may have been insufficient for complete evacuation of deep necrotic tissue and inflammatory exudate (25). Residual infectious foci may sustain local inflammation and fibroblast proliferation, potentially contributing to wide, raised, hypertrophic scars (26). Twelve patients developed wound dehiscence due to inadequate drainage and excessive exudate, with 7 progressing to secondary infection. All underwent secondary debridement followed by gauze packing and serial dressing changes until healing. However, this approach prolonged healing and, via repeated mechanical insult and local infection, contributed to tissue defects and scar formation, ultimately impairing breast aesthetics (21). These findings indicate that conventional drainage yields suboptimal wound outcomes, highlighting the need for improved management strategies (27,28). By contrast, the VSD comprehensive approach uses active negative-pressure drainage with irrigation to clear infectious material before closure, potentially reducing inflammation-driven abnormal scar formation (29,30). Uniform negative pressure may also guide granulation tissue growth and promote more orderly tissue organization (31). Delayed secondary closure was performed only after adequate wound bed preparation, a principle supported by evidence that complete wound bed preparation is a strong predictor of healing (risk ratio 2.4–4.1). A 2023 randomized controlled trial validated this approach in contaminated wounds, and breast surgery studies have confirmed its benefits in reducing complications and improving scar outcomes (32). Overall, the integrated VSD approach may facilitate removal of severe lesions while preserving and restoring the natural breast contour, which is particularly relevant to women of childbearing age, the population predominantly affected by GLM.
Although comprehensive VSD wound management showed long-term advantages, it was associated with greater short-term morbidity and healthcare resource use. Patients in the VSD group reported more severe pain (5.989±0.896 vs. 3.984±0.813, P<0.001), required more frequent dressing changes (14.862±3.638 vs. 10.730±5.074, P<0.001), and had longer incision healing times (38.081±7.007 vs. 19.048±12.681 days, P<0.001). In addition, hospital stays (9.931±1.797 vs. 8.984±2.703 days, P=0.01) and hospitalization costs (¥12,186.47±1,348.62 vs. ¥11,628.57±1,274.00, P=0.01) were significantly higher in the VSD group than in the conventional drainage group. These findings reflect the inherent complexity of VSD-based management, which entails continuous device maintenance, regular monitoring, and a prolonged wound-healing trajectory. Notably, these additional therapeutic burdens were superimposed on substantially greater baseline disease severity in the VSD comprehensive group relative to the conventional drainage group, underscoring the trade-off between enhanced clinical efficacy and increased resource utilization in this more challenging patient population.
The 4.6% recurrence rate observed in the VSD comprehensive group compares favorably with the 1.0–50.0% range reported in previous GLM surgical series (17). Our findings are consistent with recent evidence suggesting that comprehensive VSD wound management may improve outcomes in complex breast infections by promoting wound healing and reducing residual foci. However, direct comparisons are limited by heterogeneity in study design, patient selection, and recurrence definitions. To our knowledge, this is among the few studies to systematically evaluate both recurrence and breast morphology outcomes after VSD in patients with GLM, providing a more comprehensive assessment of its clinical utility.
Several limitations of this study should be acknowledged. First, the retrospective, nonrandomized design is subject to selection bias, although baseline imbalances were quantified using SMDs and sensitivity analyses were performed to adjust for confounding. Second, the relatively small sample size and limited number of recurrence events (n=17) restricted the extent of multivariable adjustment and may have reduced the power to identify independent predictors of recurrence. Third, because this was a single-center study, the generalizability of the findings to broader populations and other clinical settings may be limited. Finally, longer follow-up is needed to confirm the durability of the observed recurrence benefit beyond 2 years.
Conclusions
In conclusion, segmental mastectomy combined with comprehensive VSD wound management was associated with lower recurrence and better breast morphology than conventional drainage in patients with IGM, despite greater baseline disease severity in the VSD comprehensive group. However, these long-term advantages were accompanied by greater short-term morbidity and higher healthcare costs. Accordingly, the decision to use VSD should be individualized according to disease severity, patient preferences, and available resources. Prospective multicenter studies with longer follow-up are warranted to validate these findings and refine patient selection criteria for VSD in GLM management.
Supplementary
The article’s supplementary files as
Acknowledgments
The authors thank Dr. Huang Haiyun, department head, for assistance with clinical data collection and language editing. AI-assisted tools were used for language polishing and manuscript-formatting support only; no AI tool was used to generate research data, perform statistical analyses, or produce figures. The authors reviewed and approved the final manuscript and take full responsibility for its content.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Gansu Provincial Hospital (No. 2025-528). The requirement for individual informed consent for this retrospective analysis was waived.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0372/rc
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0372/coif). The authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0372/dss
References
- 1. Li H, Lin C, Huang M, et al. Epidemiology and antimicrobial resistance of pathogens in granulomatous mastitis: A multicenter study in Guangdong, China. iScience 2025;28:113629.. 10.1016/j.isci.2025.113629 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Rashaan MI, Hamid BS, Rashid MJ, et al. Clinical and epidemiological criteria for idiopathic granulomatous mastitis among women in Sulaymaniyah, Iraq: A retrospective study. Medicine (Baltimore) 2026;105:e49343.. 10.1097/MD.0000000000049343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Yazkan C, Dinç F, Tekin L, et al. Idiopathic Granulomatous Mastitis Mimicking Breast Cancer in a Male Patient: Case Report. Int J Surg Pathol 2026;34:231-5.. 10.1177/10668969251361177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Anousha K, Jahanbin B, Ardalan FA, et al. The diagnostic dilemma of idiopathic granulomatous mastitis with an emphasis on histopathologic findings. Diagn Pathol 2025;21:8.. 10.1186/s13000-025-01739-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Yuan QQ, Xiao SY, Farouk O, et al. Management of granulomatous lobular mastitis: an international multidisciplinary consensus (2021 edition). Mil Med Res 2022;9:20.. 10.1186/s40779-022-00380-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bretelle M, Miquel C, Mhamdi N, et al. Clinical features and therapeutic management of idiopathic granulomatous mastitis. Clin Exp Dermatol 2026;51:461-3.. 10.1093/ced/llaf493 [DOI] [PubMed] [Google Scholar]
- 7. Liu J, Lin WY, Wang QZ, et al. Unveiling the enigma: understanding the complex disease course of granulomatous mastitis and seeking ways to shorten it. Gland Surg 2025;14:646-57.. 10.21037/gs-2024-557 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Zhang F, Li C, Wu X, et al. Non-inferiority of minimally invasive rotational cutting in granulomatous mastitis treatment: a comparative trial. Sci Rep 2025;15:728.. 10.1038/s41598-024-79778-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Vercoe J, Sedaghat N, Brennan ME. Intralesional Steroid Injections for Management of Granulomatous Mastitis: A Systematic Review of Treatment Protocols and Clinical Outcomes. Breast J 2025;2025:2592366.. 10.1155/tbj/2592366 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Fleischmann W, Strecker W, Bombelli M, et al. Vacuum sealing as treatment of soft tissue damage in open fractures. Unfallchirurg 1993;96:488-92. [Article in German]. [PubMed] [Google Scholar]
- 11. Qian H, Lei T, Hu Y. Negative pressure wound therapy versus gauze dressings in managing open fracture wound of lower limbs: A meta-analysis of randomized controlled trials. Foot Ankle Surg 2022;28:1120-8.. 10.1016/j.fas.2022.03.012 [DOI] [PubMed] [Google Scholar]
- 12. Huang T, Huang H. Application effect of vacuum sealing drainage combined with nano-silver dressings in the treatment of non-healing wounds after surgery for open limb fractures. J Orthop Sci 2026;31:207-12.. 10.1016/j.jos.2025.05.012 [DOI] [PubMed] [Google Scholar]
- 13. Chinese Society of Pathology. Breast Pathology Group, Tumor Pathology Committee of China Anti-Cancer Association . Chinese expert consensus on the pathological diagnosis of granulomatous lobular mastitis (2024 version). Zhonghua Bing Li Xue Za Zhi 2024;53:996-1004. [Article in Chinese].. 10.3760/cma.j.cn112151-20240612-00382 [DOI] [PubMed] [Google Scholar]
- 14. Jafarov R, Aliyev A, Babazade I, et al. Personalized Treatment Outcomes in Idiopathic Granulomatous Mastitis: A Retrospective Study of Ninety-Two Patients. Eur J Breast Health 2026;22:270-8.. 10.4274/ejbh.galenos.2026.2025-9-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Zhang M, Pu D, Meng F, et al. Alterations in Signaling Pathways and Therapeutic Strategies of Traditional Chinese Medicine in Granulomatous Lobular Mastitis. J Inflamm Res 2025;18:9185-97.. 10.2147/JIR.S535195 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kapoor NS, Ryu H, Smith L, et al. Presentation and Management of Granulomatous Mastitis in the United States: Results of an American Society of Breast Surgeons Registry Study. Ann Surg Oncol 2024;31:7396-404.. 10.1245/s10434-024-15714-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zhou Y, Xu L. Clinical efficacy of different methods for treatment of granulomatous lobular mastitis: A systematic review and network meta-analysis. PLoS One 2025;20:e0318236.. 10.1371/journal.pone.0318236 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Shang B, Zhang T, Liu C, et al. Analysis of three treatment methods for granulomatous lobular mastitis: a retrospective study in a single center. Front Oncol 2025;15:1588836.. 10.3389/fonc.2025.1588836 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Lin Y, Hong Z, Xie W, et al. Ultrasound-guided microwave ablation versus conventional surgery for Granulomatous Lobular Mastitis: A two-center retrospective comparative stud. PLoS One 2026;21:e0351776.. 10.1371/journal.pone.0351776 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Chiu HS, Huang TS, Chen CT, et al. Temporal Regulation of Early-Stage Cytokine Expression in Diabetic Wound Healing Under Negative Pressure Wound Therapy. Int J Mol Sci 2025;26:4634.. 10.3390/ijms26104634 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Milcheski DA, Clivatti GM, Santos Junior RA, et al. Effectiveness of negative-pressure wound therapy with instillation compared to standard negative-pressure wound therapy and traditional gauze layer dressing for the treatment of acute traumatic wounds: A randomized controlled trial. J Plast Reconstr Aesthet Surg 2025;100:208-18.. 10.1016/j.bjps.2024.11.005 [DOI] [PubMed] [Google Scholar]
- 22. Ioannidis O, Anestiadou E, Zapsalis K, et al. Tailored Negative Pressure Wound Therapy with instillation in diabetic, hypertensive, and obese patients-when guideline treatment is not enough: a case series and a proposal for the ANSWER score. World J Emerg Surg 2025;20:66.. 10.1186/s13017-025-00605-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Siribumrungwong B, Chantip A, Noorit P, et al. Comparison of Superficial Surgical Site Infection Between Delayed Primary Versus Primary Wound Closure in Complicated Appendicitis: A Randomized Controlled Trial. Ann Surg 2018;267:631-7.. 10.1097/SLA.0000000000002464 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Syafawani N, Ali MI. Carbuncle management: skin-sparing saucerization followed by delayed primary closure is a new treatment modality: a case series. J Med Case Rep 2025;19:618.. 10.1186/s13256-025-05309-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Stelmar J, Smith SM, Chen A, et al. Procedures Never Explained in Textbooks: How to Correctly Convert a Closed-Suction Drain to a Closed-Gravity Drain, and How to Correctly Remove a Closed-Suction Drain Off Suction. J Surg Res (Houst) 2022;5:419-22.. 10.26502/jsr.10020236 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Yang N, Zhang H, Zhang Y, et al. Bacterial colonization contributes to pathological scar formation via the regulation of inflammatory response. J Transl Med 2025;23:569.. 10.1186/s12967-025-06585-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fan CW, Chen PH, Jhou HJ, et al. Negative pressure wound management in perineal wound status post abdominoperineal resection and extralevator abdominoperineal excision: a meta-analysis and trial sequential analysis. Int J Colorectal Dis 2023;38:73.. 10.1007/s00384-023-04353-5 [DOI] [PubMed] [Google Scholar]
- 28. Zhang J, Xu R, Qian J, et al. Closed incision management with negative pressure wound therapy: a systematic review and meta-analysis of orthopedic trauma outcomes. BMC Musculoskelet Disord 2025;26:991.. 10.1186/s12891-025-08986-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Li L, Jia H. Small-incision, point-by-point debridement combined with double-cannula negative-pressure irrigation and drainage for granulomatous mastitis: A single-center retrospective cohort study. J Int Med Res 2026;54:3000605261462010.. 10.1177/03000605261462010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hu S, Xiao J, Nie X, et al. Vacuum sealing drainage combined with continuous irrigation for the treatment of oral and maxillofacial abscesses-a retrospective study. BMC Oral Health 2025;25:257.. 10.1186/s12903-025-05634-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Wu M, Matar DY, Yu Z, et al. Continuous NPWT Regulates Fibrosis in Murine Diabetic Wound Healing. Pharmaceutics 2022;14:2125.. 10.3390/pharmaceutics14102125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Carter MJ, Snyder RJ, Zarbiv KD, et al. The Correlation Between Wound Bed Preparation and Wound Closure in Venous Leg Ulcers: A Post Hoc Analysis of the ChronEx Multicenter Randomized Controlled Trial. Adv Wound Care (New Rochelle) 2026;15:32-40.. 10.1177/21621918251366644 [DOI] [PubMed] [Google Scholar]