Our study showed that the treatment management and outcomes in patients with LGM present a complex clinical scenario. With the rising global incidence of LGM cases, the complexities in managing patients become increasingly apparent, particularly among special groups such as those with underlying medical conditions. In our study, corticosteroids, as the primary treatment, was administered to a majority of patients, showing varying degrees of effectiveness and tolerability. Among those treated with corticosteroids, a notable proportion (42%) experienced complete resolution of symptoms, while others faced relapse (11%) or showed no improvement (24%). Additionally, a significant number of patients discontinued corticosteroids due to reasons such as poor compliance, adverse medication effects, or pregnancy. Previous research has presented varying rates of treatment success among LGM patients receiving corticosteroids either alone or combined with other treatment modalities. A meta-analysis involving 358 LGM patients treated with oral corticosteroids between January 1, 2010, and December 31, 2015, showed complete remission rates ranging from 30.8–100%, with recurrence rates ranging from 0–46.2%. The pooled estimates for complete remission and recurrence rates of corticosteroids were 71.8% [95% CI (confidence interval) 67.1%, 76.3%] and 20.9% (95% CI 9.2%, 16.1%), respectively. When oral corticosteroids were combined with surgery, the estimated complete remission and recurrence rates were 94.5% (95% CI 88.9%, 98.3%) and 4% (95% CI 1.5%, 8.4%), respectively (16). In another systematic review and meta-analysis covering corticosteroids in LGM involving 559 patients up to May 21, 2019, the recurrence rate in the corticosteroids -only group was 17.7%. The relative risk and risk difference of recurrence in the steroid-only group compared with the surgery-only group were 2.99 (95% CI 0.28–31.33) and 0.14 (95% CI − 0.01–0.30), respectively, showing no significance. Additionally, the relative risk of recurrence in the corticosteroids-only group compared to the combined therapy of corticosteroids plus surgery was 6.13 (95% CI 0.41–81.62), again showing no significance. However, the risk of recurrence in the steroid-only group was significantly higher than in the corticosteroids plus surgery group (risk difference: 0.28, 95% CI 0.11–0.44) (16). Administering corticosteroids as an adjunctive therapy prior to surgical interventions in patients with LGM has also demonstrated effectiveness in reducing inflammation (17). A study involving 156 patients who received 5 days of in-patient steroid therapy before excisional surgery revealed that only 5.1% of them experienced recurrence, a significantly lower rate compared to those who solely received prednisone (p < 0.01) (18). In addition to systemic steroid administration, the utilization of topical steroid application and intralesional injections has also shown promising results (19, 20).
Adverse effects attributed to corticosteroid in a considerable proportion of patients who discontinued treatment in our study underscore the need for close monitoring and management of side effects in LGM treatment. It also highlights the difficulties linked with prolonged corticosteroid treatment in managing LGM, emphasizing the need for additional research into alternative steroid-sparing therapies for this condition.
In situations where corticosteroid therapy proved insufficient or was not well-tolerated, MTX was prescribed as an alternative treatment option. Oral MTX is commonly employed as a second-line therapy for patients with IGM who either do not respond to or cannot tolerate steroids (12). While MTX showed promising results in half of our patients, approximately 10% experienced recurrences during follow-up. Moreover, MTX-related side effects were reported, leading to treatment discontinuation in some other cases. Adverse reactions to MTX have been documented in previous studies, ranging from none to 18.2% (21–23). However, in our cases, side effects were observed in only about 5% of patients. Retrospective analyses have reported remission rates ranging from 75–100% and recurrence rates between 12.5% and 15.8% with MTX monotherapy, typically administered at doses of 7.5 to 25 mg per week over an average treatment duration of 8.5 to 15 months (13, 23, 24). The choice between oral and subcutaneous MTX administration depends on individual patient factors and tolerability.
Literature indicates that MTX is often used in conjunction with corticosteroids. Combining prednisone with MTX at doses of 5 to 10 mg per week has led to remission rates ranging from 58.5–100%, with relapse rates varying from zero to 28.6% (21, 22, 24–28). Discrepancies in outcomes of our patients compared to previous studies may be attributed to the inclusion of treatment discontinuation as a separate category within treatment outcomes. While some patients who discontinued treatment experienced improvement due to the self-limiting nature of the disease, they were classified as treatment discontinuation cases rather than being included in the remission group.
A small subset of patients in our study eventually required surgical intervention, such as lumpectomy, indicating the refractory nature of the disease in certain cases. Surgical resection is often regarded as the most efficient treatment approach in LGM with the shortest resolution time. A meta-analysis revealed that surgical excision significantly increased the complete remission rate compared to steroid therapy (P = 0.0003). However, the study reported no significant difference in effectiveness between observation and surgical intervention for early LGM patients with mild symptoms (relative risk (RR) = 0.78, 95% CI [0.55, 1.11], P = 0.17) (29). Another meta-analysis reported a recurrence rate of 22.5% for various surgical procedures including drainage, excision, and lumpectomy (30). Moreover, surgical procedures carry potential complications such as impaired wound healing, fistula or abscess formation (ranging from 4.7–30.0%), scarring, and breast asymmetry. These factors have led surgical intervention to be considered an alternative approach for patients who do not respond to medical therapy rather than being the primary management option (31–37).
Upon our final follow-up conducted after a median duration of five years, we discovered that over 90% of LGM patients achieved complete resolution of their symptoms, irrespective of the treatment approaches utilized. Similarly, a meta-analysis revealed that 95.5% of LGM patients with mild symptoms experienced spontaneous resolution (29). While secondary outcome of the study, assessed through long-term follow-up, demonstrated a high rate of complete resolution among our patients, underscoring the self-limiting nature of LGM, it's important to note that we did not conduct a comparative analysis to assess the severity of mastitis at presentation and the cosmetic outcomes between patients receiving medical treatment and those managed conservatively. Moreover, a proportion of patients continued to experience symptoms or exhibited incomplete resolution during the study follow-up, emphasizing the chronic and variable nature of LGM.
There has been considerable debate regarding the optimal approach to managing LGM, whether it should be actively treated or managed through observation, and the most effective treatment modality for those who require intervention: medical or surgical. LGM is categorized into four stages based on disease progression and clinical presentation: (1) self-limited stage, (2) congestive swelling stage, (3) abscess formation stage, and (4) complex refractory stage. At the self-limited stage, and watchful waiting through clinical examination has been proposed as a reasonable strategy. During this stage, symptoms may spontaneously resolve or remain stable for months or even years (38). In our study, 11 (7%) cases initially underwent expectant management and showed improvement over a median period of approximately 10 months. Similarly, previous research has indicated that in patients who were observed without active treatment, symptom resolution typically occurred within a range of 5 to 14.5 months (8–10, 39–41). Davis et al. observed that delayed first childbirth was correlated with a longer duration of watchful waiting (8).
Several factors have been identified as contributors to the more persistent course of LGM. In our investigation, elevated prolactin levels and the occurrence of breast abscesses subsequent to treatment were linked to prolonged mastitis duration. Patients who developed breast abscesses post-treatment exhibited a substantially longer illness duration compared to those who did not experience such complications. The median duration of illness in this subgroup was 23 months, significantly longer than the median duration of 12 months observed in patients without abscess formation. This association underscores the clinical significance of abscess development as a marker of disease severity and complexity. Similarly, prior research has shown that the recurrence of LGM is more prevalent among those who develop recurrent abscesses during treatment (42). The prolonged illness duration in patients with abscesses may be attributed to factors such as delayed healing and the need for additional interventions such as surgical drainage.’ Hur et al. demonstrated that patients with lesions measuring 1–2 cm in diameter tended to experience a self-limited condition, while those with larger lesions (> 5 cm) were more prone to progress to breast abscess (39). Our study did not identify significant associations between other characteristics and extended illness duration. Characteristics like marital status, the use of hormonal contraceptives, a family history of breast cancer, and clinical signs such as the presence of a mass, erythema, nipple discharge, axillary lymphadenopathy, EN, and the size of the mass did not demonstrate statistically significant associations with the duration of the disease. Previous studies have indicated that purulent nipple discharge, skin lesions, bilateral disease, pain, a Body Mass Index (BMI) of ≥ 24 (indicative of overweight/obesity), and an elevated follicle-stimulating hormone (FSH)/luteinizing hormone (LH) ratio are associated with a heightened risk of recurrence (43–46). Our findings demonstrated that although patients experiencing pain, nipple retraction, fistula drainage, and bilateral involvement tended to have a longer disease duration, these differences did not reach statistical significance. While some of these factors may contribute to disease pathogenesis and presentation, they may not independently influence the duration of illness in LGM patients.
LGM can present unique challenges in individuals with comorbidities such as diabetes mellitus, hyperprolactinemia, and erythema nodosum (EN). Our study investigated the impact of these comorbid conditions on the disease course and treatment outcomes in LGM patients. Elevated levels of plasma prolactin were observed in a subset of patients, with a notable proportion exhibiting pituitary adenomas. Patients with LGM and hyperprolactinemia experienced a prolonged disease course compared to those with normal prolactin levels, emphasizing the influence of hormonal factors on disease progression. Treatment with bromocriptine or cabergoline was effective in managing hyperprolactinemia-associated LGM, highlighting the importance of addressing underlying hormonal imbalances in treatment strategies. Consistent with our findings, previous studies have also suggested an association between elevated prolactin levels and the recurrence of LGM (45, 47). Elevated prolactin levels can lead to increased milk production and accumulation within the mammary lobules, potentially causing infection or extravasation into the perilobular stroma, triggering a T-cell-mediated immune response and subsequent granuloma formation (48). Moreover, prolactin has been shown to activate the nuclear factor kappa-light-chain-enhancer of activated b cells (NF-kB) signaling pathway in mammary epithelial cells, leading to the production of pro-inflammatory cytokines such as interleukin (IL)-1, IL-6, (tumor necrosis factor) TNF-a, interferon (INF)-c, and Granulocyte macrophage colony stimulating factor (GM-CSF), which could further exacerbate inflammation and contribute to granuloma formation in the breast (49). Thus, elevated prolactin levels may exacerbate inflammation and influence the duration of recovery, underscoring the importance of routine screening for hyperprolactinemia in LGM patients (9, 50).
In our study, 12.8% of LGM patients exhibited elevated plasma prolactin levels. A meta-analysis conducted in 2023 reported a prevalence of hyperprolactinemia in 19.7% of LGM patients (99 out of 502) (30). Furthermore, 2.6% of our cases were diagnosed with prolactinoma. This finding suggests an unusually high prevalence of prolactinoma among LGM patients compared to the incidence of 60–100 cases per 1,000,000 individuals in general population (51).
Treating LGM in special patient groups, such as those with diabetes, poses notable challenges. Although diabetes has been reported in 6.2% of LGM cases, no clear association has been established between diabetes and the onset or recurrence of LGM (44). Hyperglycemia is known to result in the formation of glycosylated end products that may stimulate B-cell proliferation and cytokine release, leading to an autoimmune response in various organs, including the breasts (52). It is also suggested that persistent hyperglycemia, along with increased intermolecular cross-linkage and glycosylation, impedes collagen degradation, contributing to connective tissue accumulation in the breasts (53). In our study, diabetes was identified in 7.7% of cases, yet these patients did not exhibit a prolonged disease course or a higher recurrence rate for LGM. Consistent with our findings, a study on LGM patients scheduled for observation found no significant difference in the time to resolution between diabetic and non-diabetic patients (RR = 0.98, 95% CI [0.59–1.63], P = 0.94) (8). Additionally, another study investigating factors contributing to LGM recurrence did not find a significant association between diabetes and disease recurrence (OR = 1.38, 95% CI [0.59–3.25], P = 0.45) (44). While diabetes did not appear to significantly impact the disease course, careful management of glycemic control and potential interactions with immunosuppressive therapies are warranted in diabetic LGM patients. In our study, five diabetic patients were treated with MTX, resulting in remission for four of them. One patient did not respond to treatment, and recurrence occurred in another patient, although both cases improved with conservative management. Prior studies have also advocated for methotrexate as the preferred initial treatment for diabetic LGM patients over steroids (23).
We identified that patients with a subtype of LGM that is associated with EN exhibited more complex disease manifestations, including multiple breast masses and concurrent arthritis. Additional research has also indicated that patients with concurrent LGM and EN often exhibit more extensive breast involvement (P = 0.01) (54, 55). The term GMENA (Granulomatous Mastitis, Erythema Nodosum, Arthritis) syndrome was introduced by Parperis et al. in 2021, indicating the simultaneous presence of LGM with EN and arthritis (56). Both GM and EN show similar histopathological findings with chronic inflammation and granulomas, suggesting a shared underlying cause (56). Despite the complexity, treatment approaches involving corticosteroids and methotrexate were effective in achieving complete recovery in this subgroup of patients. Previous studies also have suggested that individuals with LGM and EN may exhibit favorable responses to systemic immunosuppression due to shared pathophysiological mechanisms linked to autoimmunity (57, 58). For instance, one study involving 11 patients with LGM and EN treated with methylprednisolone reported full recovery within 12 weeks, with no recurrence observed during the 60-month follow-up period (59). In our cohort, patients with LGM and EN who received either prednisone or MTX tended to have a longer disease duration compared to those without EN, although this contrast did not reach statistical significance. However, two previous studies demonstrated a significant association between the presence of both LGM and EN and a prolonged disease course (P < 0.001, and P = 0.005) (55, 60). In our study, half of the patients with both LGM and EN had experienced a previous episode of LGM managed with prednisolone. Several studies have indicated a significantly higher recurrence rate in LGM patients with EN compared to those without EN (42.31% vs 16.00%, P < 0.001) (60, 61). However, another study found a higher recurrence rate in the EN group, but the difference was not statistically significant (16.7% vs 6.7%, P = 0.24) (55).
Our study has limitations that warrant acknowledgment. Firstly, it was conducted at a single center, potentially restricting the applicability of our findings to broader populations. However, the inclusion of a large sample size and patients with various underlying conditions may partially address these limitations. Additionally, our study recruited patients exclusively from our institution, which could introduce selection bias. This bias might result in the overrepresentation of individuals with more severe or treatment-resistant cases of LGM, potentially inflating assessments of disease severity and treatment efficacy.